How good is the Lucky feat

[Skrum]

Apologize if this has been done before, but I got into a friendly disagreement with a fellow player because he insisted Lucky was far and above the best feat in the game. This seems obviously and trivially false to me, but whatever. Is this a topic of hot debate or something? What's the general feeling about Lucky?

[claypigeons]

Lucky is very good. Whether it's the 'best' is easily contested by circumstances.

It dies, however, offer one the chance to choose their own destiny. It specifically allows you to choose which die to use. Which means in the case of rolling at disadvantage on save, for instance, you could use a luck point, roll a third d20, and then choose which one you want...not just the lowest. That can be a significant enough boon to warrant taking it.

[Greywander]

Two things about the Lucky feat:

1. Anyone can make use of it. It doesn't matter what your build is, Lucky will give you an edge.

2. The less you need it, the stronger it becomes. Optimizers will get the most benefit out of it, as they'll only rarely need to use it. This might seem counter intuitive, but it makes it a lot easier to never get hit or never fail a save or ability check or whatever you're optimized for. The better optimized you are, the less often you'll need to use Lucky, and when you do use it the more likely you are to succeed on the reroll.

I do think people might overestimate the value of Lucky. It's definitely nice to have, and it can be quite satisfying when it saves you from a bad roll, but it doesn't really let you do anything you couldn't already do. It's a good feat to grab if you've already gotten all your other build-defining feats and boosted your primary ability score to 20.

[Unoriginal]

Apologize if this has been done before, but I got into a friendly disagreement with a fellow player because he insisted Lucky was far and above the best feat in the game. This seems obviously and trivially false to me, but whatever. Is this a topic of hot debate or something? What's the general feeling about Lucky?

It is hotly debated, yeah. Some are up all night for good fun, others are up all night to get Lucky.


More seriously, Lucky is one of those feats that mostly show up in optimization discussions or when you've already spent all the ASIs you needed/wanted and are spending the remaining ones on useful but not essential things.

I personally am not a fan, but there is no objective metric for what would constitute the "best" feat.

[Skrum]

It is hotly debated, yeah. Some are up all night for good fun, others are up all night to get Lucky.


More seriously, Lucky is one of those feats that mostly show up in optimization discussions or when you've already spent all the ASIs you needed/wanted and are spending the remaining ones on useful but not essential things.

I personally am not a fan, but there is no objective metric for what would constitute the "best" feat.

What you said and the comment above yours confirms my intuition: lucky is the "second best" feat in the game. As in, it's sweet to have, but it would hardly ever be worth taking when it comes to picking a feat or ASI.

Its apparent benefit is hugely magnified because it might turn a bad roll into a good one. That is very memorable, and thus feels extremely powerful. But the reality is that it's not nearly as strong as an ability that gives a consistent, regular benefit.

[Eldariel]

It's a great feat. It's also overrated by some (it's not literally broken) and underrated by some (it's certainly above average). It's hard to evaluate because it's one feat that takes a bit of player skill to use well - you need awareness of roll impact and relative likelihood of needing the reroll later to really make use of it. 3 uses is plenty for one day but it means you can't go around rerolling e.g. attack rolls except under highly exceptional circumstances.

Generally, what it is good for:
- Reroll saves. Especially if you're about to fail one of your strong saves against an effect that could really **** you over, go for it. Poisoned or such isn't usually worth saving (unless it's like Pit Fiend poison), but if you're failing Concentration (that matters: obviously if you're already mopping up and your Hypnotic Pattern doesn't really do much but hold one enemy in place that you're about to wail on anyways, that stray bolt or random aura dropping your Concentration doesn't matter) or like a Wis/Int/Cha-save against something but a fear effect (or if you're a martial and thus care about fear effects, those too), those are generally the kind of saves you should pay attention to.
- Reroll enemy critical hits. Normal hits aren't worth noting but some monsters have really nasty critical hits (basically anything that rolls a lot of dice, from lowly Hobgoblins to Mummies, Chasmes, Shadow Demons attacking from hiding, any higher level humanoid, Dragons with elemental bites, etc.) and you generally don't want to be tanking those if you can avoid it no matter who you are. So if you are fighting something that rolls a fistful of dice and you're about to get crit, make them try again.
- Key skill checks. If you rolled a 1 for that important Persuasion check and are about you get booted out of the kingdom by the evil vizier's advice with hounds on your tail, yes, that is absolutely a time you should reroll that Persuasion check. If the consequence of failure is just a bit of damage or whatever, no, don't burn it there but if it's a make-or-break kinda deal, all the yes.
- Initiative. There are times when you really need to go first and those times, if you roll 1-7 for your Initiative, you're probably rerolling it. Especially if you're a spellcaster, and especially if you have good bonuses like Alert Chronurgist where you, with a decent roll, can expect to go first and as a spellcaster, can land that key CC or summon or whatever gamechanging effect that turns the fight from a near certain disaster to an easy victory.


With rules like that, you generally get mileage out of that but you generally have enough rolls to get through at least most of the important rolls of the day. I use it as a surrogate for Res: Con or War Caster on many characters since practically speaking you don't expect to be failing that many Concentration-saves by just the right amount per day to need Res: Con/War Caster more than 3 times leaving you with the option of rerolling other things too if need be. Of course, Res: Con is great if you have odd Con but it's less amazing if you started with an even score.

I would look to pick it up by Tier 3 on most characters simply because the amount of saves that can **** your character over raises with level (especially the amount of harm a failed save can do - you can literally be rerolling a character or TPKing the party with a single bad roll) and nobody has reliable way to make those DC20+ saves so you really want to mitigate the chance of failure as much as you can, so as to minimise the risk of total existential failure.

On spellcasters (that just cast, don't fight), my first three feats are generally Alert, Lucky and Res: Con. Add to that +4 to main stat and that takes you to level 16 (assuming Custom Lineage or Variant Human). But if you're a fighting type, you need feats for fighting and don't have the luxury of trying to avoid dying (Fighter gets to pick these up around Tier 3 with those extra ASIs). It's still a great tanking feat in particular, because of the aforementioned ability to reroll crits.

[Skrum]

It's a great feat. It's also overrated by some (it's not literally broken) and underrated by some (it's certainly above average). It's hard to evaluate because it's one feat that takes a bit of player skill to use well - you need awareness of roll impact and relative likelihood of needing the reroll later to really make use of it. 3 uses is plenty for one day but it means you can't go around rerolling e.g. attack rolls except under highly exceptional circumstances.

Generally, what it is good for:
- Reroll saves. Especially if you're about to fail one of your strong saves against an effect that could really **** you over, go for it. Poisoned or such isn't usually worth saving (unless it's like Pit Fiend poison), but if you're failing Concentration (that matters: obviously if you're already mopping up and your Hypnotic Pattern doesn't really do much but hold one enemy in place that you're about to wail on anyways, that stray bolt or random aura dropping your Concentration doesn't matter) or like a Wis/Int/Cha-save against something but a fear effect (or if you're a martial and thus care about fear effects, those too), those are generally the kind of saves you should pay attention to.
- Reroll enemy critical hits. Normal hits aren't worth noting but some monsters have really nasty critical hits (basically anything that rolls a lot of dice, from lowly Hobgoblins to Mummies, Chasmes, Shadow Demons attacking from hiding, any higher level humanoid, Dragons with elemental bites, etc.) and you generally don't want to be tanking those if you can avoid it no matter who you are. So if you are fighting something that rolls a fistful of dice and you're about to get crit, make them try again.
- Key skill checks. If you rolled a 1 for that important Persuasion check and are about you get booted out of the kingdom by the evil vizier's advice with hounds on your tail, yes, that is absolutely a time you should reroll that Persuasion check. If the consequence of failure is just a bit of damage or whatever, no, don't burn it there but if it's a make-or-break kinda deal, all the yes.
- Initiative. There are times when you really need to go first and those times, if you roll 1-7 for your Initiative, you're probably rerolling it. Especially if you're a spellcaster, and especially if you have good bonuses like Alert Chronurgist where you, with a decent roll, can expect to go first and as a spellcaster, can land that key CC or summon or whatever gamechanging effect that turns the fight from a near certain disaster to an easy victory.


With rules like that, you generally get mileage out of that but you generally have enough rolls to get through at least most of the important rolls of the day. I use it as a surrogate for Res: Con or War Caster on many characters since practically speaking you don't expect to be failing that many Concentration-saves by just the right amount per day to need Res: Con/War Caster more than 3 times leaving you with the option of rerolling other things too if need be. Of course, Res: Con is great if you have odd Con but it's less amazing if you started with an even score.

I would look to pick it up by Tier 3 on most characters simply because the amount of saves that can **** your character over raises with level (especially the amount of harm a failed save can do - you can literally be rerolling a character or TPKing the party with a single bad roll) and nobody has reliable way to make those DC20+ saves so you really want to mitigate the chance of failure as much as you can, so as to minimise the risk of total existential failure.

On spellcasters (that just cast, don't fight), my first three feats are generally Alert, Lucky and Res: Con. Add to that +4 to main stat and that takes you to level 16 (assuming Custom Lineage or Variant Human). But if you're a fighting type, you need feats for fighting and don't have the luxury of trying to avoid dying (Fighter gets to pick these up around Tier 3 with those extra ASIs). It's still a great tanking feat in particular, because of the aforementioned ability to reroll crits.


You make a good case. I do like the idea of using it strategically like that, especially as characters get to higher levels. It's not as bad as I was thinking.

Still.....guy I'm talking to says it beat out an ASI, on a fighter, at 4th level. That seems insane to me.

[Rukelnikov]

You make a good case. I do like the idea of using it strategically like that, especially as characters get to higher levels. It's not as bad as I was thinking.

Still.....guy I'm talking to says it beat out an ASI, on a fighter, at 4th level. That seems insane to me.

Well, im more inclined to consider Lucky amongst the best feats in the game than otherwise. Point is, is not a build enabling feat, as you yourself mentioned a couple posts ago.

At 4th lvl as a fighter, you are better off taking something that will enable you to fulfill your role better, GWM, PAM, XBE, SS, Sentinel, something like that. Thing is, there are only so many feats that grant you (meaningful) support for your role, and no matter what that role is, it will almost assuredly have to make dice rolls at one point or another, even if its to make saving throws.

And thus eventually most builds will benefit from being Lucky. And IMO it's not a stretch to say that a feat which is not the priority for anyone but its very useful to everyone is the best feat in the game.

[Jerrykhor]

I mean, if you think its not that great, then nobody can change your mind.

But you're wrong about it not being better than ASI. Absolutely wrong.

[LudicSavant]

Lucky is very good. Whether it's the 'best' is easily contested by circumstances.

I concur.

Strong feat. Not "far and above the best" feat.

[Zhorn]

It specifically allows you to choose which die to use. Which means in the case of rolling at disadvantage on save, for instance, you could use a luck point, roll a third d20, and then choose which one you want...not just the lowest. That can be a significant enough boon to warrant taking it.
For Skrum I'd say take the above with a grain of salt and always check with your DM on how THEY run the feat at their tables, as the above method is not universal.

Depending on your DM; there are three different ways that interaction will go for pairing Lucky with Advantage/Disadvantage:
Lucky turns any Advantage/Disadvantage into your choice of the 3 dice results.
Resolve Advantage/Disadvantage first, they apply Lucky to the single resultant die as normal.
Apply Lucky to one of the two dice rolled for advantage/disadvantage, then apply the rules for advantage/disadvantage as normal.
Both rulings (1) and (3) are in the Sage Advice Compendium, and (2) isn't all that uncommon either.

It is still a good feat for almost any build under and of these three rulings, but how it ranks compared to other ASI options changes a bit if your DM isn't fond of ruling (1).

[Eldariel]

For Skrum I'd say take the above with a grain of salt and always check with your DM on how THEY run the feat at their tables, as the above method is not universal.

Depending on your DM; there are three different ways that interaction will go for pairing Lucky with Advantage/Disadvantage:
Lucky turns any Advantage/Disadvantage into your choice of the 3 dice results.
Resolve Advantage/Disadvantage first, they apply Lucky to the single resultant die as normal.
Apply Lucky to one of the two dice rolled for advantage/disadvantage, then apply the rules for advantage/disadvantage as normal.
Both rulings (1) and (3) are in the Sage Advice Compendium, and (2) isn't all that uncommon either.

It is still a good feat for almost any build under and of these three rulings, but how it ranks compared to other ASI options changes a bit if your DM isn't fond of ruling (1).

Yeah, this is an interaction to clear up since RAW is stupid on this front. I personally run it as follows:
Advantage + Lucky: Roll 3, pick best.
Disadvantage + Lucky: Roll 3, pick middle.

This feels most consistent and they can all just be rolled at once without regard for which is the Lucky die and which the other dice. But this is by no means RAW.





You make a good case. I do like the idea of using it strategically like that, especially as characters get to higher levels. It's not as bad as I was thinking.

Still.....guy I'm talking to says it beat out an ASI, on a fighter, at 4th level. That seems insane to me.

There are some edge cases where I'd consider that, like if I were running a Fighter who wants to tank more than anything. The ability to reroll enemy crits is huge for a frontliner: you are unlikely to get hit by non-crits so the reroll is likely to result in a miss removing a huge spike of damage 3/day, while also giving you cushion for getting disabled by save-or-****-yous. Indeed, I'd rate Lucky as a "great tank feat" for Fighters. Now, builds that want to/can afford to take a tank feat at 4, that's a bigger question but e.g. Spear & Board Vuman PAM Duelist Battlemaster has their only relevant feat already and can easily afford Lucky on 4 though of course this means they put resources into defense instead of offense (PAM Battlemaster has plenty of tools to be disruptive already).

But that's not strictly wrong even though Fighter is much defined by their offense. Could take Heavy Armor Master on 4, Crusher [using Quarterstaff] on 6, Lucky on 8 and be a really disruptive tanky mofo with still 18 Str. Then on 12 Res: Wis, +2 Str on 14, etc. By and large though, generally Fighter needs to work on their offense first since that's literally their only contribution. Battlemaster is a bit of an exception on that front. Of course, as Battlemaster you have the lure of those extra dice from feats, which do add a bunch to your SR contributions, which does raise the opportunity cost of picking up these feats instead.

[Hytheter]

It depends on your play environment. I largely play one-shot discord text games where more than one fight in a session is unsual, and Lucky truly shines there. Being able to reroll on demand can drastically change the path of an encounter, and being able to do it three times in one fight is pretty fantastic.

It definitely has the advantage that it's generically good - anyone can use it.

[diplomancer]

Any argument about a feat that says "this is the absolute best use of an ASI under any circumstance" is wrong.

That said, Lucky is probably the best "all-around" feat, since every single character benefits from it. For that very reason, more specialized feats are often better/more urgent.

[longgone]

Yeah, this is an interaction to clear up since RAW is stupid on this front. I personally run it as follows:
Advantage + Lucky: Roll 3, pick best.
Disadvantage + Lucky: Roll 3, pick middle.

RAW is quite clear. It's roll 3, pick whichever you want. This is also clarified in Sage Advice which also gives alternatives for DMs who wants to play it differently.

Either way, it's a good feat but not unbalanced.

[Eldariel]

RAW is quite clear.

I know. I said RAW is stupid.

[Contrast]

I mean, if you think its not that great, then nobody can change your mind.

But you're wrong about it not being better than ASI. Absolutely wrong.

I would never take Lucky over an ASI to a main stat. Even as a variant human I'm unconvinced you don't have at least 1 better option to pick.

I think Lucky is probably the most overrated feat in the game which is mostly surprising because it is actually one of the better feats - some people just seem to think its super broken and it really really isn't, its just generically and solidly good. I've played lots of games with people with Lucky and its never been more impactful than, say, PAM or GWM/SS. Most builds would potentially have space for Lucky by level 16. At level 12 though pretty much no-one should have picked it up if they were trying to fully optimise for my money.

It is also worth noting that the power of the feat changes heavily depending how many encounters/situations for normally get into in a given day. 1-2 and it will seem more powerful, 6-8 and it'll seem less powerful.

Personally I think I'd almost always have some random flavour feat that I'd be more interested in pursuing than adding Lucky.

[Waazraath]

More seriously, Lucky is one of those feats that mostly show up in optimization discussions or when you've already spent all the ASIs you needed/wanted and are spending the remaining ones on useful but not essential things.

I personally am not a fan, but there is no objective metric for what would constitute the "best" feat.

This. And even after haven taking essential feats/stat increases for a build, I think it has pretty fierce competition from a feat like Alert.

[Eldariel]

I would never take Lucky over an ASI to a main stat. Even as a variant human I'm unconvinced you don't have at least 1 better option to pick.

As a caster, I've done it more than once. I usually haven't ended up regretting it (though preparing 1 spell less is rough). Unlike martials, casters don't need feats for their basic proficiency so their ASIs are free and while casting stat improvements are nice, Lucky goes a long way towards helping you succeed on Counterspell checks, Dispel checks, Concentration checks, Initiative, etc. so it's got quite significant perks (notably in things that are otherwise very hard to buff like Counterspell-checks; +1 from casting stat is far lesser a buff than the reroll from Lucky and you aren't casting Counterspell often enough for Lucky's limited uses to come into play). While being limited to at most 3 encounters per day does suck, most of those encounters tend to be quite trivial so you don't end up burning Lucky except in harder ones and in those you tend to get good value for your money.

[longgone]

I know. I said RAW is stupid.

And that it needed clearing up. It doesn't, it is already clear.

[Eldariel]

And that it needed clearing up. It doesn't, it is already clear.

It needs clearing up, not because RAW is unclear but because RAW is stupid. You should check with your DM if they follow RAW even in cases where it makes no sense; these are the types of things DMs commonly rule however they want and thus you should ask about it before it comes up and without assuming they will follow RAW because, at least based on anecdotal experience, RAW-followers seem to be in the minority on this topic.

[Lunali]

Lucky is the best feat in the same way that con is #1 for stats. It's rarely the best feat for any given character, but it's always a good feat, making it better overall than any other feat.

As for whether it's better than an ASI, that depends on your criteria. If you're just looking for the largest boost in effectiveness, an ASI will probably be better as it applies to every roll with that stat. If you're looking for something that makes the game more fun, feats (lucky included) are generally better.

[MoiMagnus]

Lucky's power also vary depending on your day's length.
If you make only 10 d20 rolls a day, Lucky is going to be significantly stronger than if you make 100 of them.

[shipiaozi]

Lucky is a really great feat, but nowhere close to CE, PAM, Inspiring Leader or -5/+10

[msfnc]

I rarely reject a pc build that’s brought to my table. One of those rare times occurred recently, when a player showed up with a halfling divination Wizard, with lucky and second chance. I thought it was a cool build, but I didn’t let them play it. Portents, Lucky, Luck, and Second Chance all working together is a bit much. They named the pc ‘Faqyu Dierolls’. They expected the rejection, and had a backup pc built. With Lucky and Elven Accuracy.

If you just want to reroll all the dice, go play Yahtzee or Savage Worlds...

[Scarlet Knight]

I feel it is the best feat, so much so I refused to take it so not to ruin my game.

What if the DM made all his NPCs lucky?

Imagine a high level fighter is rolling to hit you, & you need him to miss or you're put out of the battle? You're ok ! Oh but wait! He's lucky...so...roll... you go down.

You get the drop on a Mind Flayer...oh sorry...he's lucky & goes first.

The wizard toying with your party fails his save! He's held! Nope, try again.

Nothing is more versatile and can ruin everybody's day in more ways than Luck.

[diplomancer]

I feel it is the best feat, so much so I refused to take it so not to ruin my game.

What if the DM made all his NPCs lucky?

Imagine a high level fighter is rolling to hit you, & you need him to miss or you're put out of the battle? You're ok ! Oh but wait! He's lucky...so...roll... you go down.

You get the drop on a Mind Flayer...oh sorry...he's lucky & goes first.

The wizard toying with your party fails his save! He's held! Nope, try again.

Nothing is more versatile and can ruin everybody's day in more ways than Luck.

Lucky on enemy NPCs is specially overpowered for the same reason that it's overpowered in a 5 minute adventuring day campaign; after all, from the NPC perspective, it almost always IS a 30 seconds adventuring day! Your examples clearly show that, specially the one about the enemy fighter; common opinion about the feat is that it is good UNLESS you do things like using it to redo a regular attack.

Of course, that says very little about its power in a regular campaign, except as a reminder, yet again, that 5e is not balanced around a 5 minutes adventuring day.

[jas61292]

I think Lucky is both an annoyingly good feat, and also super overrated.

Feats are a massive investment. You are giving up a +2 to your ability scores to get it. A +1 to all of your most common rolls is far and away more powerful that a couple of rerolls. Its not even particularly close, in my opinion. What's more, when compared with other feats, Lucky does not let you do anything new or different. You are investing an ASI into simply being slightly more likely to succeed at what you already do, but often not to the same extent as simply boosting your ability scores.

Additionally, Lucky, like all long rest based things, often gets vastly overrated online where a lot of people admittedly do not follow the games encounter guidelines. Yes, it is absolutely a more powerful ability when you have very few encounters per day. But if you do stick to the guidelines, having a single reroll every other encounter is very, very underwhelming.

That being said, yes, Lucky is still good. While in the abstract it is probably weaker than many alternatives, you choose wh,en it comes up, which does increase its power. Furthermore, it is applicable all the time with any character, which is very nice. I think the people who have been saying it is always one of the strongest feats, but never the actual best choice, are hitting the nail on the head. It is never bad to have, but it is rarely going to be optimal, outside very high levels where other feats and ASIs have already been acquired.

Now, I also referred to it as "annoyingly good," and the reason for that is that, in my personal opinion, a feat so bland and flavorless, that adds next to nothing non-mechanical to the game, does not deserve to be as good as it is. Despite its name, Lucky does very little to make a character actually feel all that lucky. Sure, its a "I won't fail that crucial save" button, but that does not translate well to making the character feel actually lucky. It is super common for someone without the feat to roll better on average than someone with it (even after accounting for rerolls). What's more, there is no particular reason that any successes from Lucky rerolls represent anything more lucky than just succeeding on it normally. I mean... how do you get lucky on a concentration save? ("I was about to lose concentration, but then, suddenly.... I didn't." Yeah, OK). To me, lucky is what you are when you accidentally push the right level to solve a puzzle, or mindlessly trip over valuable treasure. It's not hitting with the occasional extra attack, or only taking a normal hit instead of a crit.

Simply put, I don't like Lucky. And I don't think it is a top choice for anyone. But it is never a bad choice, and I would never tell someone they screwed up by picking it.

[Keravath]

Lucky is at best a second tier feat. In most cases, an ASI to bump stats can be significantly better than Lucky over the long term.

Some folks seem to think Lucky is overpowered for some reason ... I have no understanding of that viewpoint since in my experience, even most optimizers won't bother with Lucky until level 16 or 19 when they don't have anything better to take.

However, Lucky is a feat that will work for any character build, this makes it a common choice once all the actually useful feats/ASI for a particular build have been used.

Lucky has a couple of good applications
1) Mitigate a critical hit by an opponent
2) Attempt to make an important save or skill check that you have already failed

The problem with saves and skill checks is that if it is an easy roll then most of the time the character makes it the first time and Lucky is an insurance policy for the low probability fail events - and Lucky can still fail on the second attempt.

On the other hand, for difficult skill checks/saves then the odds of failing the first time are pretty high and once you have failed, the odds of failing the second time are the same as the first - so even with Lucky you still probably fail.

I think the main reason some folks have a different opinion of Lucky is a combination of observational/memory bias and actual Luck. Successes with Lucky are far more memorable and even more memorable the more important a die roll. However, if you fail then a second fail with Lucky isn't particularly memorable. The second is that if you happen to actually get lucky then the second roll can allow for changing a failure into a success. Some players can be very lucky so that almost every time they re-roll with Lucky they succeed. This is very memorable but it is also a side effect of the player actually being lucky and not the feat if they succeed regularly on low probability die rolls.

The three uses/day of Lucky is extremely limiting except in one scenario ... if the D&D game has only one encounter/fight in an adventuring day and the players know that then Lucky becomes more powerful since the player can use all three uses in one combat. This could potentially swing things since a Lucky paladin could use the feat to land extra smites for example, or a rogue to succeed on a sneak attack or similar - applications of the feat which would normally be excluded if you were saving it for important saves and crit mitigation - however, if players know there is only one fight then they figure that they might as well use it.

Finally, I've seen several characters with the Lucky feat - I have it on one of mine, I am DMing a character using it and I have played with a small number of folks who had it. For the most part, it has been very weak in practice - it either doesn't get used waiting for an important event to use it on or it gets used up and isn't available when that time arrives. Perhaps 2-3 times in all the uses has it made any difference - usually, if they failed the first time then the odds are they will fail the second. That is just my personal experience with the feat - it is generally a weak feat that makes a good last choice for almost any character.

[JNAProductions]

I feel it is the best feat, so much so I refused to take it so not to ruin my game.

What if the DM made all his NPCs lucky?

Imagine a high level fighter is rolling to hit you, & you need him to miss or you're put out of the battle? You're ok ! Oh but wait! He's lucky...so...roll... you go down.

You get the drop on a Mind Flayer...oh sorry...he's lucky & goes first.

The wizard toying with your party fails his save! He's held! Nope, try again.

Nothing is more versatile and can ruin everybody's day in more ways than Luck.

I'll echo Diplomancer a bit in response to this.

If your DM has a pool of three Luck Points for use per Long Rest on any NPC, that'd probably be fine. If every NPC has their own pool of three Luck Points, for a decent chunk of them, they won't make more than three d20 rolls, because they die fast.

To the OP and general thrust of the thread: Lucky is good. There's no character it's bad on-some make more use of it than others, but no character would find it useless. Is it the best, though? Not in my opinion. It's dang good, but certainly not always better than an ASI or more focused feat.

[Kurt Kurageous]

I rarely reject a pc build that’s brought to my table. One of those rare times occurred recently, when a player showed up with a halfling divination Wizard, with lucky and second chance. I thought it was a cool build, but I didn’t let them play it. Portents, Lucky, Luck, and Second Chance all working together is a bit much. They named the pc ‘Faqyu Dierolls’. They expected the rejection, and had a backup pc built. With Lucky and Elven Accuracy.

If you just want to reroll all the dice, go play Yahtzee or Savage Worlds...

I built and played something very close to that (before Second Chance existed). What it was best at was enchantment spells that were guaranteed to work. I loved low portent rolls because I could make the strong +6, +8 save bonus irrelevant.

Not saying it was OP as things can ignore enchantment, but at times it felt like Suggestion (walk away from me until the sun sets!) was the best spell ever to remove a source of conflict from the scene.

[KorvinStarmast]

But you're wrong about it not being better than ASI. Absolutely wrong. That kind of assertion is a great way to ensure that one's post is poorly received.
It depends on your play environment.
Yes.

It definitely has the advantage that it's generically good - anyone can use it. Even better.

Our AT has it (he picked it at level 8) and he's just now beginning to get a 'feel' for it. (He's not a fan of reading internet forums that focus on optimization). He's learning by doing.

And he's doing great. :smallsmile:

[Keravath]

I know. I said RAW is stupid.

Actually, I don't see it that way. The Lucky feat is supposed to make the character lucky - give them a better chance to succeed. Personally, I don't see any issue with the character being the MOST lucky when the odds are stacked against them. The point of being lucky is to succeed when everyone else would likely fail and I think the option to choose the best die out of 3 when rolling with disadvantage + Lucky tends to reflect that. (Luke and the death star in Star Wars is probably the classic example :) ).

Mechanically, you can argue about whether it is overpowered but I find that it just doesn't come up enough to be an issue - how many times during an adventuring day does a character roll with disadvantage and really want to use one of their rare Lucky re-rolls on it? Most disadvantage is on to hit rolls for one reason or another which is a substandard use of Lucky anyway.

So, because it is uncommon and seems to me to be thematically appropriate, I have no problem with changing disadvantage to choosing the best die of the three when using Lucky.

[JNAProductions]

Actually, I don't see it that way. The Lucky feat is supposed to make the character lucky - give them a better chance to succeed. Personally, I don't see any issue with the character being the MOST lucky when the odds are stacked against them. The point of being lucky is to succeed when everyone else would likely fail and I think the option to choose the best die out of 3 when rolling with disadvantage + Lucky tends to reflect that.

Mechanically, you can argue about whether it is overpowered but I find that it just doesn't come up enough to be an issue - how many times during an adventuring day does a character roll with disadvantage and really want to use one of their rare Lucky re-rolls on it? Most disadvantage is on to hit rolls for one reason or another which is a substandard use of Lucky anyway.

So, because it is uncommon and seems to me to be thematically appropriate, I have no problem with changing disadvantage to choosing the best die of the three when using Lucky.

The main issue I see with that is being able to intentionally give yourself Disadvantage and burn a Luck Point to make it triple Advantage.

[Keravath]

The main issue I see with that is being able to intentionally give yourself Disadvantage and burn a Luck Point to make it triple Advantage.

True enough :) ... they could close their eyes to make an attack and "trust in their Luck".

On the other hand, a character doesn't know how many uses of Luck they have in a day so if the player wanted to have a character that closes their eyes for all their attacks throughout the day then I'd say go for it. If they only wanted to do it when they had Luck points left, I'd have a little chat about the difference between character knowledge, player knowledge and role playing.

[Amechra]

Part of me thinks that Lucky should've just been the default for PCs (using the "disadvantage is roll 3 dice and pick the middle one" house rule). If you also made it so that luck points came back slowly, you could even replace Inspiration with "you get a luck point".

[MoiMagnus]

Actually, I don't see it that way. The Lucky feat is supposed to make the character lucky - give them a better chance to succeed. Personally, I don't see any issue with the character being the MOST lucky when the odds are stacked against them. The point of being lucky is to succeed when everyone else would likely fail and I think the option to choose the best die out of 3 when rolling with disadvantage + Lucky tends to reflect that. (Luke and the death star in Star Wars is probably the classic example :) ).

Mechanically, you can argue about whether it is overpowered but I find that it just doesn't come up enough to be an issue - how many times during an adventuring day does a character roll with disadvantage and really want to use one of their rare Lucky re-rolls on it? Most disadvantage is on to hit rolls for one reason or another which is a substandard use of Lucky anyway.

So, because it is uncommon and seems to me to be thematically appropriate, I have no problem with changing disadvantage to choosing the best die of the three when using Lucky.

I see this argument, but it makes Lucky somewhat ridiculous. Want to use lucky? First you need to find a way to have a disadvantage (close your eyes while attacking, insult the king before rolling for persuasion, etc), and then use lucky to transform that disadvantage in a bonus.

I'd be totally fine with Lucky worded as overwriting any disadvantage/advantage, and making you roll 3 take 1 whatever the situation. But giving bonuses to negative circumstances can lead to absurd behaviours, which are very fun in parodic RPGs, but might not be adequate in more serious campaigns.

EDIT:


On the other hand, a character doesn't know how many uses of Luck they have in a day

Why, though? Not much more absurd than knowing how many action surge you still have.
The character might have "a bad feeling about it" or "a good feeling about it" depending on how many luck point they still have.
I don't see why the character wouldn't know how much luck they still have, with the same level of precision that they know how much vital energy they still have.

[kbob]

Lucky is the best feat in the same way that con is #1 for stats. It's rarely the best feat for any given character, but it's always a good feat, making it better overall than any other feat.

I concur with this. Lucky often won’t be taken early on with most builds as many will want the advantage of their build to come online ASAP. Though there are a few that take it earlier as their thought is that survivability is more important than the “build” (this philology is more common in 3.5 as feats are more plentiful but these individuals do exist in 5e). It’s just a different gaming philosophy. Regardless, it typically shows up on the character at some point. Unless you’re a melee build you won’t have PAM or GWF. Mostly casters take resilient/con and only casters take war caster. However, most players that min/max (and even many that don’t) end up taking lucky at some point. More take lucky than I know take alert. And A lot of people pick up alert. The thing with lucky is that it helps anyone who takes it and will augment any build that you are making that requires dice rolls (I can’t think of a build that doesn’t use dice rolls). Alert will come up at initiative (which is big) and occasionally with surprise (which is good when it arises). Lucky can give you advantage on initiative if you want and disadvantage on any attack made against you including a surprise (which there are other ways around negating surprise). In addition to all the things mentioned from above posters: super advantage when you should have disadvantage (confirmed by Crawford when his tweets were official rules interpretations), disadvantage on an enemy attack and almost disadvantage if they had advantage (if your roll was lower than their top... this can cancel nasty crits), crit fishing when you NEED it against the boss, making sure that save/suck doesn’t fail, and I know I already stated it but making disadvantage into super advantage (instead of taking lowest you take the highest of 3 dice)... in addition to all these, you get 3 uses per day! It’s an incredible feat.
I mean why not take it? Except for flavor reasons, there no reason not to. At some point anyway. ANYONE who takes it will get A LOT of mileage out of it. You are cheating the one part of the game no one has control over (including the DM), your dice.
Used on specific builds, it goes from incredible to just plain stupid. Give it to a divination wizard (you’re all set). Want more? Make him a yuanti and never worry about failing a save again, especially if there is a paly in the group. You can focus your efforts on attacks or making the enemy fail it’s save if your portent rolls were low (which personally I love getting 1s over 20s with my Divwiz).
Have a paly and really need to increase your odds of critting? Take drow, elven accuracy accuracy, dip hex for 2, take devil sight, lucky. Attack with 4 dice/attack (up to 3 times) and crit on 19/20 against BBEG when you cast darkness (or find another way to get advantage). Take PAM at some point so you can take all three lucky rolls in 1 round for your attack+BA. That’s 3 attacks but rolling 12d20 do so. I think a 19 or 20 may creep in there.
Those are just some builds that work around lucky. But again ANYONE who will be rolling dice will get a lot of mileage out of it. And the more skilled you are at the game (as someone else mentioned) the better it will be for you. Which makes an already good feat even better. A noob using it will get a lot out of it. A skilled player who knows the optimal times to use it will change the outcome of the game. Imagine if the whole party had it? It is the best feat in the game because anyone will benefit greatly from it, it augments any build, it has the power to affect chance (the one thing no one can control) whenever you want even after dice rolls (but before outcome), and it has the potential to change the outcome of any encounter (social/combat/puzzle/trap). If it requires a roll, lucky is here to help. And you can do it 3 times a day!

[KorvinStarmast]

Actually, I don't see it that way. The Lucky feat is supposed to make the character lucky - give them a better chance to succeed. Personally, I don't see any issue with the character being the MOST lucky when the odds are stacked against them. Yep.

Most disadvantage is on to hit rolls for one reason or another which is a substandard use of Lucky anyway. I have seen that too.

So, because it is uncommon and seems to me to be thematically appropriate, I have no problem with changing disadvantage to choosing the best die of the three when using Lucky. Only three chances per Long Rest. The only problem is when the 5 minute adventure day is the norm, but that causes a whole lot of problems all around.

I'd be totally fine with Lucky worded as overwriting any disadvantage/advantage, and making you roll 3 take 1 whatever the situation. Yeah, it also makes for some cool moments in play during really tight battles.

[MrStabby]

Apologize if this has been done before, but I got into a friendly disagreement with a fellow player because he insisted Lucky was far and above the best feat in the game. This seems obviously and trivially false to me, but whatever. Is this a topic of hot debate or something? What's the general feeling about Lucky?

Lucky... ranges from tier 2 feat to being insanely good.

The way I think about it is to think of all your die rolls it can impactin a typical adventuring day. Line them all up in order of impact. So that save against hold person that means you lose concentration on your Wall of Force that was keeping a pit fiend off your back whilst you handled the other half of the fight might be at the top... an attack roll that you don't care about missing being at the bottom. It is the top end of this scale that you care about. Those things that really put your party at risk - the massive criticals, failed saves, botched intitiative rolls, failled counterspell attempts. These are the things that count.

So this depends on your DM. If your DM just bum-rushes you with a load of mooks regularily and you don't face saves vs feeblemind or banishment or real problem spells. If they don't use enemies that can crit for 60% of your life total or more with one hit and so on then it is less good. It all depends on how much can hang on one die roll.

The other point about adventuring day lenth still holds, but I would say in my experience that adventuring day length has less impact on lucky than how much can be hung on specific die rolls.

[Reynaert]

What if the DM made all his NPCs lucky?

Then the DM would have to determine how many other encounters these NPCs have had, where they could have used up those Luck points.

Oh, but NPC's don't have many encounters per day you say? They just wait around for the players to encounter them?

Well, then there's your answer: It's very unfair to give NPC's the Lucky feat, or more generally it's very unfair to build NPC's like they're player characters.

[diplomancer]

Then the DM would have to determine how many other encounters these NPCs have had, where they could have used up those Luck points.

Oh, but NPC's don't have many encounters per day you say? They just wait around for the players to encounter them?

Well, then there's your answer: It's very unfair to give NPC's the Lucky feat, or more generally it's very unfair to build NPC's like they're player characters.

Yeah, someone made an interesting suggestion about that; 3 luck points for the DM to give to NPCs over the course of an adventuring day; maybe makes more sense if the enemies du jour have a hivemind, like a Mindflayer Colony; the lucky mindflayers!

[heavyfuel]

I'd say it's pretty strong, but, weirdly balanced. You see, there are usually better feats to be gotten, so it's not usually worth the opportunity cost.

It's always better to max your main ability score first, so usually you're not taking feats before level 8. And if you are taking a feat, you're probably taking something that's integral to your build (PAM, EA, Warcaster, GWM, SS, etc).

Later on, when one might take Lucky, it's in direct competition with other "secondary" feats (Alert, Inspiring Leader, Chef, etc). As someone who loves playing casters, and who has played both a Wizard with Lucky and a Wizard with Alert, I'll say Alert was consistently giving me an edge, especially since a lot of days we'd take a long rest and I'd still have 1 or 2 lucky dice.

Lucky is amazing if you rolled stats and rolled them well, which might free up a couple of ASIs.

[Angelalex242]

When I take Lucky, it's usually my level 19 ASI.

By Tier 4, you're going to face rolls you don't want to fail. A lot.

[KorvinStarmast]

I'd say it's pretty strong, but, weirdly balanced. You see, there are usually better feats to be gotten, so it's not usually worth the opportunity cost.

It's always better to max your main ability score first, so usually you're not taking feats before level 8. And if you are taking a feat, you're probably taking something that's integral to your build (PAM, EA, Warcaster, GWM, SS, etc). Alert is an undervalued feat. You don't really 'need' to max your primary ability score until 12, but I tend to lean toward 'max it by 8' if I have a spell caster due to how spell save DCs operate.

But think about this for a moment:
vHuman barbarian, Alert feat at level 1.
+5 to initiative
Go with Unarmored Defense, and get Dex to 16 or 18.
Now get to level 7 with Feral Instinct.
Advantage on Initiative rolls. And add +5.

barb goes first a lot, and if they rage in the first round (advantage on Athletics checks, right, when raging) they can knock prone, attack, blah blah as a first move.

It's a concept I am hoping to try out soon, but it will take a one shot since none of our party Barbarians have the alert feat yet ...

[quindraco]

I'd say it's pretty strong, but, weirdly balanced. You see, there are usually better feats to be gotten, so it's not usually worth the opportunity cost.

It's always better to max your main ability score first, so usually you're not taking feats before level 8. And if you are taking a feat, you're probably taking something that's integral to your build (PAM, EA, Warcaster, GWM, SS, etc).

Later on, when one might take Lucky, it's in direct competition with other "secondary" feats (Alert, Inspiring Leader, Chef, etc). As someone who loves playing casters, and who has played both a Wizard with Lucky and a Wizard with Alert, I'll say Alert was consistently giving me an edge, especially since a lot of days we'd take a long rest and I'd still have 1 or 2 lucky dice.

Lucky is amazing if you rolled stats and rolled them well, which might free up a couple of ASIs.

It's tougher as a wizard, since many spells require a visible target - I don't know offhand how many attack spells do. But certainly e.g. you can just consume a luck point on a scorching ray (which doesn't require a visible target) by shutting your eyes as you cast. The only class I can think of with a legit excuse to not run out of luck points is rogues, since disadvantage turns off sneak attack even after luck.

[Angelalex242]

It also depends on race.

If you're playing a +2 to your primary stat race...and then set that stat to 17...

The idea is to take a half feat to get your stat to 18, then get your stat to 20.

Alternatively, the vuman is also going to take a half feat at level 1 (probably), but then needs to spend 4 and 8 on pure ASIs to get to 20.

[quindraco]

It also depends on race.

If you're playing a +2 to your primary stat race...and then set that stat to 17...

The idea is to take a half feat to get your stat to 18, then get your stat to 20.

Alternatively, the vuman is also going to take a half feat at level 1 (probably), but then needs to spend 4 and 8 on pure ASIs to get to 20.

VHumans are dead post-Tasha's - the custom lineage is them but better. A custom lineage with a half-feat is 18 at L1, and can be 20 by L4.

[Eldariel]

VHumans are dead post-Tasha's - the custom lineage is them but better. A custom lineage with a half-feat is 18 at L1, and can be 20 by L4.

Nah, there are plenty of cases where you want 16/16 and a full feat (since full feats are better than half feats) over 15/18 and a half-feat.

[KorvinStarmast]

Nah, there are plenty of cases where you want 16/16 and a full feat (since full feats are better than half feats) over 15/18 and a half-feat. vHUman Monks come to mine, among others.

[heavyfuel]

It's tougher as a wizard, since many spells require a visible target - I don't know offhand how many attack spells do. But certainly e.g. you can just consume a luck point on a scorching ray (which doesn't require a visible target) by shutting your eyes as you cast. The only class I can think of with a legit excuse to not run out of luck points is rogues, since disadvantage turns off sneak attack even after luck.

So you burned two resources (lucky die and spell slot) to maybe deal low damage with a single ray to a single target, or you burned FOUR resources (3 lucky dice and a spell slot) to maybe deal considerable damage.

Oh, and also everyone gets advantage against you because you decided to blind yourself for the round (unless the DM allows you to close and open your eyes at specific moments for absolutely no cost. I don't think there's any RAW about that, so it is entirely up the DM, unlike the "triple advantage" RAW)

I honestly don't see the problem. I'll still go for max ability score on any caster. Extra save DC is just that good.

[Rukelnikov]

So you burned two resources (lucky die and spell slot) to maybe deal low damage with a single ray to a single target, or you burned FOUR resources (3 lucky dice and a spell slot) to maybe deal considerable damage.

Oh, and also everyone gets advantage against you because you decided to blind yourself for the round (unless the DM allows you to close and open your eyes at specific moments for absolutely no cost. I don't think there's any RAW about that, so it is entirely up the DM, unlike the "triple advantage" RAW)

I honestly don't see the problem. I'll still go for max ability score on any caster. Extra save DC is just that good.

What? You blinked TWICE this turn? That will cost you your bonus action.

[Tanarii]

It dies, however, offer one the chance to choose their own destiny. It specifically allows you to choose which die to use. Which means in the case of rolling at disadvantage on save, for instance, you could use a luck point, roll a third d20, and then choose which one you want...not just the lowest. That can be a significant enough boon to warrant taking it.
I've never bought into this.

IMO either:
- you get to choose one of the dice in disadvantage and lucky can replace it, then you resolve disadvantage by comparing to the other.
- you get to resolve disadvantage first to get the lowest of two dice, and then use the lucky dice to replace it.

IMo the first is the way it should be resolved. But the second still makes way more sense than "pick any of the three".

[kbob]

I've never bought into this.

IMO either:
- you get to choose one of the dice in disadvantage and lucky can replace it, then you resolve disadvantage by comparing to the other.
- you get to resolve disadvantage first to get the lowest of two dice, and then use the lucky dice to replace it.

IMo the first is the way it should be resolved. But the second still makes way more sense than "pick any of the three".

Crawford said, when his tweets/posts were official, that With disadvantage, RAW, you do indeed choose between the 3 dice. However, at his own table he plays that you drop the highest of the original 2 dice then choose between the remainder and the third die. I would run it the same but that is not RAW.

[Tanarii]

Crawford said, when his tweets/posts were official, that With disadvantage, RAW, you do indeed choose between the 3 dice. However, at his own table he plays that you drop the highest of the original 2 dice then choose between the remainder and the third die. I would run it the same but that is not RAW.
I disagree with Crawford and that it's not a valid interpretation of RAW.

[Zhorn]

I disagree with Crawford and that it's not a valid interpretation of RAW.

I very much dislike the cheese of getting to choose between all three of the dice as it negates Disadvantage completely (or Advantage for the opponent), but the wording of the feat does not contain a line limiting the Lucky dice to only interacting with a single d20


Whenever you make an attack roll, an ability check, or a saving throw, you can spend one luck point to roll an additional d20. You can choose to spend one of your luck points after you roll the die, but before the outcome is determined. You choose which of the d20s is used for the attack roll, ability check, or saving throw.

It's not a case of Super (Dis)Advantage being the clear intent, but it is still a valid point to arrive at by not clarifying either a single dice interaction or the specifics of how it should act under Advantage/Disadvantage.

Personally I don't allow that at my tables and make it clear to my players that Lucky is applied against a single dice before Advantage/Disadvantage is resolved, but Super (Dis)Advantage is still technically valid :smallyuk:

[Tanarii]

I very much dislike the cheese of getting to choose between all three of the dice as it negates Disadvantage completely (or Advantage for the opponent), but the wording of the feat does not contain a line limiting the Lucky dice to only interacting with a single d20



It's not a case of Super (Dis)Advantage being the clear intent, but it is still a valid point to arrive at by not clarifying either a single dice interaction or the specifics of how it should act under Advantage/Disadvantage.

Personally I don't allow that at my tables and make it clear to my players that Lucky is applied against a single dice before Advantage/Disadvantage is resolved, but Super (Dis)Advantage is still technically valid :smallyuk:
It is valid, but it's equally valid to arrive at the lucky dice roll/choice being involved either before the 2nd die is thrown for disadvantage, or after it's been thrown and one of them determined to be lower and removed.

[quindraco]

It is valid, but it's equally valid to arrive at the lucky dice roll/choice being involved either before the 2nd die is thrown for disadvantage, or after it's been thrown and one of them determined to be lower and removed.


No, it isn't. 5E has genuine, actual rules for timing, you know.

Let's take an attack roll the Lucky person is rolling (for example). By definition, Lucky happens after the attack roll - the Lucky person rolls the attack roll, then, after the roll is made, chooses whether or not to spend a point on an additional die. That's right there in the feat.

Meanwhile, as PHB p173 makes clear, when you have advantage or disadvantage on a roll, the second die is rolled as part of that roll. In our example, the second die is rolled as part of the attack. Then, after both dice are rolled, the Lucky die may be rolled or not. There is no legal way to roll the Luck die before the adv/disadv die - that would be cheating.

I also disagree with JC, but only because your second listed option, unlike the first (which is illegal, as explained above), is the actual RAW, and I don't understand how JC messed that up. The mechanics are perfectly clear, RAW, on how Luck and Disadvantage work:

1: Roll 2 dice at once, and pick the lowest. That's how a disadvantaged attack roll works. Until you pick the lowest, the attack roll is not complete - the definition of a disadvantaged attack roll is roll two dice, pick lowest.
2: Now that the roll is over, you have one die showing a number. That's the result of your roll. You can spend a luck point or not to roll an additional die.

In other words, JC is clearly violating the RAW when he has you roll Luck before the attack roll is completed - Luck can only be legally rolled after the attack roll, at which point there is no way to have two dice showing faces. You only have two dice showing faces during the attack roll, which is before Luck can be legally rolled.

[Snails]

I've never bought into this.

IMO either:
- you get to choose one of the dice in disadvantage and lucky can replace it, then you resolve disadvantage by comparing to the other.
- you get to resolve disadvantage first to get the lowest of two dice, and then use the lucky dice to replace it.

IMo the first is the way it should be resolved. But the second still makes way more sense than "pick any of the three".

Yours both seem like more sensible mechanics for applying Luck to Disadvantage in a way that honors the flavor of both.

"Pick any of the three" definitely does not make sense to me, in terms of intent. Yet it definitely seems to be the best reading of the text as written, IMHO.

[Snails]

If Lucky is so awesome, how many Vumans take it at level 1?

It is a second tier feat, in the same bucket are those feats very rarely taken instead of a primary stat ASI but compete well against bumping Con. In contrast, there are definitely first tier feats that can be argued as worth not boosting the primary stat for at level 4/6/8.

[MrStabby]

If Lucky is so awesome, how many Vumans take it at level 1?

It is a second tier feat, in the same bucket are those feats very rarely taken instead of a primary stat ASI but compete well against bumping Con. In contrast, there are definitely first tier feats that can be argued as worth not boosting the primary stat for at level 4/6/8.

Well it is also possible that it is a feat awesome later on but not great in tier 1.

And it can be powerful but devoid of flavour and players want and can have both.


With respect to the rules on Lucky wrt advantage and disadvantage I am also pretty cool with the more lenient use. I figure "lucky" describes success where failure would be expected, rather than success where success was expected. Occasionally doing the unlikely seems OK.

[Ogre Mage]

Very good. Other feats might be more optimal for specific character types, but Lucky is good for everyone. Once my main attribute is at 20 Lucky is almost always a feat I consider seriously.

[Corran]

Very good. Other feats might be more optimal for specific character types, but Lucky is good for everyone.
A feat being good for everyone would matter towards how we rate it if there were some additional restrictions in place regarding how we select feats. For example, if all pc's have to pick the same feat, then I could see why we could say that it matters that lucky is good enough for everyone. Of course lucky is not equally good for every pc, and if it's not good enough for my pc then it wont become any better just because it's working wonderfully for an ally.

=============

Tanks and skill monkeys can make very good use of it, depending on various factors (all of these have been explained in great detail already throughout this whole thread), but something I didn't notice anyone saying, is how lucky can also be good as a risk mitigator. Lowering the risk in a high risk/high reward strategy will often be very much needed when trying to convince yourself that this strategy is worth pursuing. This does not necessarily have to do with optimization (eg we could be talking about a high risk/ low reward strategy), but when you really want your character to do X, and number crunching indicates that this is not the best idea, having lucky may skew the results (or our past experience) enough for us to do it.

Tarzan: So you are saying that I cannot do it?
DM: I am saying there is no point. Between your move and your cunning action, you have enough movement in this round to get down the ladder just like everyone else did, and find yourself next to the cultists.
Tarzan: I am just saying, it's much cooler to enter combat swinging from a rope. I din't buy the grappling hook for nothing. And you said that if I pull it off I get advantage on my attack. So, I am aiming my grappling hook at the chandelier and with my bonus action I...
DM: Sure, the DC is, uhm..., 17 (DM is not in favor of silly antics). If you fail you'll take some falling damage and end your turn prone somewhere over here (*points at the map)
Tarzan: Great! What do I roll?
DM: Whatever you want.
Tarzan: Ooh, I am going to use my acrobatics then! (*he said after glancing at his one remaining lucky die).

Lucky: Making your silly ideas slightly less silly on average since 2014.

[Trafalgar]

Let's try a scenario to see what's best. For this scenario, I will use a fighter who just made 4th level. The fighter has Archery Fighting Style, a 16 Dex, and a longbow. The choice is between an ASI in Dex, the Sharpshooter Feat, and the Lucky feat. We will use the standard Orc as the opponent for an AC of 13 and 15hp

ASI in Dexterity
To Hit: +4(18Dex) +2(Prof) +2(Archery Fighting Style)= +8 or any roll of a 5 or higher. %chance of hitting = 80%. Crit chance = 5%.
Avg Damage: 4.5 + 4(18Dex) = 8.5 damage
Avg Damage per attack: .8 * 8.5 = 6.8 hp
Crit Damage: 4.5 * .05 = .23hp
Total = 7.03hp average damage per attack
It will take 2.13 attacks to kill the Orc.

Sharpshooter Feat
To Hit: +3(16Dex) +2(Prof) +2(Archery Fighting Style) -5(Feat)= +2 or any roll of a 11 or higher. %chance of hitting = 50%. Crit chance = 5%.
Avg Damage: 4.5 + 3(16Dex) + 10(Feat) = 17.5 damage
Avg Damage per attack = .5 * 17.5 = 8.75 hp
Crit Damage: 4.5 * .05 = .23hp
Total = 8.98hp average damage per attack
It will take 1.67 attacks to kill the Orc.

Lucky Feat
To Hit: +3(16Dex) +2(Prof) +2(Archery Fighting Style) = +7 or any roll of a 6 or higher. %chance of hitting = 75%. Crit chance = 5%. With advantage from the lucky feat, %chance of hitting = 94%. Crit chance = 9.8%
Avg Damage: 4.5 + 3(16Dex) = 7.5 damage
Avg Damage per attack = .94 * 7.5 = 7.1 hp
Crit Damage: 4.5 * .098 = .44hp
Total = 7.55hp average damage per attack
It will take 1.98 attacks to kill the Orc.

So in this very specific scenario, Sharpshooter is hands down the best choice because you will probably kill the orc in two attacks. When you add in the other Sharpshooter benefits (ignoring cover and long range) it's a no brainer as the first choice.

I would argue that an ASI in Dexterity is your second best choice. The ASI may do less damage per attack but only a .5hp difference. You only get to use Lucky for 3 attacks that's only a 1.5hp bonus once per long rest which kind of sucks. Add in the dexterity bonus's to AC, saves, and skills and I think an ASI is clearly better than Lucky. And you could use Lucky for the Orcs attack rolls against you but that +1 AC from the Dex ASI works all time vs disadvantage on 3 attacks.

Please check over my math and let me know if I did something wrong. It was never my strongest subject in school.

[Lunali]

If Lucky is so awesome, how many Vumans take it at level 1?

It is a second tier feat, in the same bucket are those feats very rarely taken instead of a primary stat ASI but compete well against bumping Con. In contrast, there are definitely first tier feats that can be argued as worth not boosting the primary stat for at level 4/6/8.

People take vhuman because they want a specific feat to make their build work. Lucky is never that feat.

Lucky is a feat that is good enough to be competitive with (though usually slightly worse) an ASI for every character. It is rarely the best option, but always a good option. This makes it #1 overall for the same reason con is the #1 stat overall. Also, the ASI may do more for your effectiveness, but feats (including lucky) tend to be more fun to play.

[Eldariel]

Let's try a scenario to see what's best. For this scenario, I will use a fighter who just made 4th level. The fighter has Archery Fighting Style, a 16 Dex, and a longbow. The choice is between an ASI in Dex, the Sharpshooter Feat, and the Lucky feat. We will use the standard Orc as the opponent for an AC of 13 and 15hp

ASI in Dexterity
To Hit: +4(18Dex) +2(Prof) +2(Archery Fighting Style)= +8 or any roll of a 5 or higher. %chance of hitting = 80%
Avg Damage: 4.5 + 4(18Dex) = 8.5 damage
Avg Damage per attack = .8 * 8.5 = 6.8 hp per attack
It will take 2.2 attacks to kill the Orc.

Sharpshooter Feat
To Hit: +3(16Dex) +2(Prof) +2(Archery Fighting Style) -5(Feat)= +2 or any roll of a 11 or higher. %chance of hitting = 50%
Avg Damage: 4.5 + 3(16Dex) + 10(Feat) = 17.5 damage
Avg Damage per attack = .5 * 17.5 = 8.75 hp per attack
It will take 1.7 attacks to kill the Orc.

Lucky Feat
To Hit: +3(16Dex) +2(Prof) +2(Archery Fighting Style) = +7 or any roll of a 6 or higher. %chance of hitting = 75%. With advantage from the lucky feat, %chance of hitting = 94%
Avg Damage: 4.5 + 3(16Dex) = 7.5 damage
Avg Damage per attack = .94 * 7.5 = 7.1 hp per attack
It will take 2.1 attacks to kill the Orc.

So in this very specific scenario, Sharpshooter is hands down the best choice because you will probably kill the orc in two attacks. When you add in the other Sharpshooter benefits (ignoring cover and long range) its a no brainer as the first choice.

I would argue that an ASI in Dexterity is your second best choice. The ASI may do .3 less damage per attack but either way, you probably need 3 attacks to kill the Orc. You only get to use Lucky for 3 attacks while the dexterity bonus is permanent. Add in the dexterity bonus's to AC, saves and skills and its clearly better than Lucky.

Please check over my math and let me know if I did something wrong. It was never my strongest subject in school.

It's worth noting that attack rolls are generally a colossal waste of rerolls (high frequency low impact roll), so this doesn't really showcase the feat at its best.

[Trafalgar]

It's worth noting that attack rolls are generally a colossal waste of rerolls (high frequency low impact roll), so this doesn't really showcase the feat at its best.

True, but its hard to make an apples to apples comparison between a combat feat and a non combat feat. Let's compare it by the frequency it benefits the player:

Dexterity ASI: The character will benefit every time they attack with a finesse/ranged weapon. The AC bonus will help every time they are attacked. They receive a bonus to all Dex saves and skill checks.
Sharpshooter: The character will benefit whenever they make an attack with a ranged weapon.
Luck: The character will benefit 3 times per long rest.

In this case, I feel Dexterity ASI is the best. I understand what you are saying, Lucky helps with that roll you really need to make, like when a failed saving throw means death. But if an event like that doesn't happen between two Long Rests, the ASI is better.

I am not saying Lucky is a bad feat, but I would only take it after my prime attribute is at 18 or 20 and I have all the other feats for my build. For me, the question is whether the Lucky feat is better than a Resilient (Wisdom) feat for a non-Wisdom character?

[Warder]

I think Lucky is very, very good. But more importantly, it's anti-fun. We've banned it at our table because we realized all it does is prevent interesting situations from happening. There's other stuff that does that too - you could argue that every time you boost your defenses that has a similar effect - but there's something fundamentally different about seeing an unfavorable roll that would potentially lead to disaster get rerolled, over and over again. Amusingly enough, it was the player who was using Lucky at the time who first brought it up, and he described it as feeling like "he'd picked a feat designed for people who hate losing". After a quick discussion, we decided to just get rid of the feat and he picked something else instead.

[Corran]

True, but its hard to make an apples to apples comparison between a combat feat and a non combat feat. Let's compare it by the frequency it benefits the player:

Dexterity ASI: The character will benefit every time they attack with a finesse/ranged weapon. The AC bonus will help every time they are attacked. They receive a bonus to all Dex saves and skill checks.
Sharpshooter: The character will benefit whenever they make an attack with a ranged weapon.
Luck: The character will benefit 3 times per long rest.

In this case, I feel Dexterity ASI is the best. I understand what you are saying, Lucky helps with that roll you really need to make, like when a failed saving throw means death. But if an event like that doesn't happen between two Long Rests, the ASI is better.

I am not saying Lucky is a bad feat, but I would only take it after my prime attribute is at 18 or 20 and I have all the other feats for my build. For me, the question is whether the Lucky feat is better than a Resilient (Wisdom) feat for a non-Wisdom character?
Think of it like this. If the encounter could become a lot more dangerous if a character fails a few key rolls, and you cannot find a better way around it so you have to take your chances, then lucky is probably better than one of that character's other ASI's/feats (primary ability boosts included). That's easy enough to discuss for a few given encounters, but dscussing it for the course of a whole campaign is more tricky (simply because we dont know what we will be facing). For example, while I would lean for the boost in my primary score, I would probably include lucky by level 8 for a paladin that is solo tanking.

Lucky is a poor substitute for the resilient (or the warcatser) feat when you've got no other bonuses applying. When it comes to saves, it's generally best to pick it if your saves are already good enough.



I think Lucky is very, very good. But more importantly, it's anti-fun. We've banned it at our table because we realized all it does is prevent interesting situations from happening. There's other stuff that does that too - you could argue that every time you boost your defenses that has a similar effect - but there's something fundamentally different about seeing an unfavorable roll that would potentially lead to disaster get rerolled, over and over again. Amusingly enough, it was the player who was using Lucky at the time who first brought it up, and he described it as feeling like "he'd picked a feat designed for people who hate losing". After a quick discussion, we decided to just get rid of the feat and he picked something else instead.
It's interesting how different people can see the same thing in a different light. For me it's the exact opposite from what you are describing. Having the safety net of a reroll has made me have my characters try things I would otherwise be wise enough not to try. I dont see this being different than any other feature of the game though. Same holds true for any kind of bonus, be that for attack or damage, mobility, or whatever else. Maybe the problem for you was that you stacked lucky on top of other bonuses and that made the challenge go away for certain characters trying certain things?

[Tanarii]

Amusingly enough, it was the player who was using Lucky at the time who first brought it up, and he described it as feeling like "he'd picked a feat designed for people who hate losing". After a quick discussion, we decided to just get rid of the feat and he picked something else instead.
The term is risk-averse. It's a feat for people who are risk-averse. Either they don't do exceptionally risky things without a safety net, or they worry about normal risk catching them off guard.

[Eldariel]

The term is risk-averse. It's a feat for people who are risk-averse. Either they don't do exceptionally risky things without a safety net, or they worry about normal risk catching them off guard.

It's also for things you can't boost otherwise. It makes Counterspell, Dispel Magic, Contact Other Plane, etc. way more reliable. Thus its optimal user is obvious.

[MrStabby]

The term is risk-averse. It's a feat for people who are risk-averse. Either they don't do exceptionally risky things without a safety net, or they worry about normal risk catching them off guard.

I am not sure it is risk aversion. I have seen peopme with the feat still take huge risks, and take on more risk over the course of the day it just means that whilst dice are left the chances of having a reward worth the risk are higher.

[kbob]

For those saying that “you take the attack roll THEN you make your lucky roll”, arguing from a timing stance, the feat does not say not do that.
“You have 3 luck points. Whenever you make an attack roll, ability check, or saving throw, you may spend 1 luck point to roll an additional d20. You can use this ability after the original roll, but before the outcome is revealed. You choose which of the d20s is used for the attack roll, ability check, or saving throw.”
“Whenever you MAKE the attack roll... roll an ADDITIONAL d20. You CAN use this ability after the original roll...You CHOOSE WHICH of the d20s”.
It does not say nor suggest that it is an effect after the attack roll. It is part of an in addition to the rolls. You can, if you wish, wait to see what the other roll(s) was/were before you choose to add your additional (meaning in addition to the original roll(s)) roll. But that is another perk of the feat. Choosing between the 3 with adv/Dadv is RAW.
It is a very versatile feat that anyone (not just certain builds) can benefit from. This alone makes it powerful. As others stated, it’s best value (typically) comes from making important saves. You can make a good case just on its versatility that it’s the best feet. However, add in the super advantage from disadvantage and it’s OP. So much so that people ban it or have discussions about how it cannot be RAW because they can’t believe it would be that powerful. It is the best feat.

[Captain Panda]

What you said and the comment above yours confirms my intuition: lucky is the "second best" feat in the game. As in, it's sweet to have, but it would hardly ever be worth taking when it comes to picking a feat or ASI.

I'd consider lucky the "best" feat generally speaking because there is no build that it will not help with. If you aren't sure what to take and don't have lucky, take lucky. The only times it should not be the feat you take is it you have a build-specific feat that you have to have to make a character work. Sharpshooter and great weapon master come to mind for that. But those, while better for their particular builds, are not as widely applicable.

Lucky will save your character's life in a clutch situation, and does so fairly reliably if used well. It dramatically outperforms a simple ASI in that sense. It also has the benefit (as has been pointed out earlier in the thread) of having amplified value for a character who is optimized already. If your character has extremely high armor and the enemy crits you, you can lucky to undo that and make them miss. Against a high CR target with fewer, bigger attacks that is enormous. Fail a key concentration save despite having a high modifier? Lucky it. Get a low initiative in a fight you desperately need to go first in? Lucky it. Fail a saving throw that might kill you? Lucky it. Fail to counterspell the enemy caster? Lucky it.

This isn't a matter of confirmation bias and only remembering the hits and not the misses; oftentimes a character built to do something well can know that if they use lucky they are pretty likely to succeed on something instead. That's enormous.

There may be some disagreement about how valuable the feat is for a couple of reasons, though.
1. Danger level.
2. Adventuring day length.

If your campaigns tend to be safer, such that failing a couple key die rolls won't actually kill your character, sure, lucky is probably a waste. Lucky is there as an emergency button, it's wasted if you just use it to reroll attacks. Also, if your campaign runs lots and lots of small encounters lucky loses some of its luster.

That said, despite what some people heatedly insist on forums, I actually think most campaigns run adventuring days with 3-5 fights, not 6-8. That's just the reality from what I've actually seen, and part of why lucky feels so good to take.

[Ogre Mage]

Some anecdotal evidence -- it is rare that I have played in a campaign with level 16-20 characters. But in all three cases there were multiple PCs with the Lucky feat.

But the vast majority of campaigns I played in never reached that far. They ended somewhere between levels 9-12. And in those games Lucky was considerably rarer. It would pop up occasionally but in the majority of games no one had it.

This seems to suggest Lucky is a "second best" feat for the large majority of PCs. Once you have maxed out your main stat and possibly taken the one other feat which is an optimal fit for your specific character type (Sharpshooter for an archer, etc.), Lucky starts to look very attractive.

[Ogre Mage]

True, but its hard to make an apples to apples comparison between a combat feat and a non combat feat. Let's compare it by the frequency it benefits the player:
I am not saying Lucky is a bad feat, but I would only take it after my prime attribute is at 18 or 20 and I have all the other feats for my build. For me, the question is whether the Lucky feat is better than a Resilient (Wisdom) feat for a non-Wisdom character?

Well of course this is dependent on how much your DM is targeting wisdom saves. But in general, at low levels I think Lucky is unquestionably superior. As you go up in levels and proficiency bonus increases, Resilient starts to gain. Someone better at math than I could perhaps suggest when the tipping point happens.




That said, despite what some people heatedly insist on forums, I actually think most campaigns run adventuring days with 3-5 fights, not 6-8. That's just the reality from what I've actually seen, and part of why lucky feels so good to take.

3-5 fights per adventuring day has been my experience as well.

[Rukelnikov]

Well of course this is dependent on how much your DM is targeting wisdom saves. But in general, at low levels I think Lucky is unquestionably superior. As you go up in levels and proficiency bonus increases, Resilient starts to gain. Someone better at math than I could perhaps suggest when the tipping point happens.

Its not as simple as just checking some odds, which is what's generally done to calc DPR or general character resiliency against different kinds of attacks. Lucky is a feat where the player is in control, whether the feat is better or worse than resilient(wis) is gonna depend on how you are using it.

For example, one could say, if im lvl 4 with +1 Wis and no proficiency in its saves I have a 50% chance to make a DC 12 Wis save, with Resilient(Wis) I would have +3 (or +4), and that would leave me with a 60% (or 65%) chance of making my save. Lucky would instead give me a 75% chance but only 3 times a day (and even then, its not "direct", since you would trigger a second 50% roll after knowing the first one failed). The thing is, if your character is failing crucial rolls more than 3 times a day , I don't even know how he/she made it to lvl 4.

Bottom line, the feat requires understanding of what you have it for in order to be useful, same as someone with SS or GWM should know when to power attack and when not to, but those feats are a little bit more forgiving, because while missing an attack against a powerful foe may be very important, if you make it thru the fight you haven't lost that resource for the next fight. Lucky is more unforgiving in that every use of it should be well spent in order to make it worth the investment of a feat. And as many have pointed out, increasing character survivability either by making saves or negating crits, is one of the ways in which it shines the most.

[Cybren]

The term is risk-averse. It's a feat for people who are risk-averse. Either they don't do exceptionally risky things without a safety net, or they worry about normal risk catching them off guard.

The better way to frame it is that Lucky is something you put in your game when you want to encourage players to take risks. Honestly kinda just feel like "luck points" should be how the games generic "inspiration" mechanic works and then you can get rid of the feat.

[Tanarii]

The better way to frame it is that Lucky is something you put in your game when you want to encourage players to take risks. Honestly kinda just feel like "luck points" should be how the games generic "inspiration" mechanic works and then you can get rid of the feat.Looking back at what I wrote, I was wrong anyway. People who "don't do exceptionally risky things without a safety net" are sane, not risk averse. :smallyuk::smallamused:

IMO it is a better description of people that want a safety net for normal risks though, as opposed to "people who hate losing"

[strangebloke]

It's a very very good defensive feat but most people build for offense so it only gets picked up later. For a point of comparison look at another good defensive feat, resilient. Resilient will often boost a single saving throw modifier by somewhere between 3 and 5 depending on your level which.... well, which only makes a difference about 1/4 of the time you roll that save (less if you have disadvantage.) but Lucky will functionally add a buff to whichever saving throw you need to make, when you need it. It will make you better at WIS saves, but it will make you better at CON saves too, and also INT saves.

Any character building purely for defense would have a tough time justifying not having it by level 8 or so.

On offense its way weaker.

[Eldariel]

I am not saying Lucky is a bad feat, but I would only take it after my prime attribute is at 18 or 20 and I have all the other feats for my build. For me, the question is whether the Lucky feat is better than a Resilient (Wisdom) feat for a non-Wisdom character?

Depends on the DC. Reroll is worth a lot when you have a decent chance of making the save already. This is why Lucky works as a Res: Con substitute for casters because the base check is DC10 so you're over 50% to make it anyways and only need to make sure you don't roll poorly. Same goes for Wis: it's a workable replacement early on (say, 1-10) but later on you do need Res: Wis. However, it's still a near-necessary complement later on (max Wis-save bonus you can get without specific roll bonuses is +11, and many high CR enemy saves are in the range of 20+ and can instantly end your character) unless you're Fighter and already have Indomitable (which does the same job though worse, forcing you to use the new roll but that's not so relevant so often).

Classes on which Lucky is great defensively:
- Any caster
- Rogue
- Ranger
- Barbarian
- Paladin (seriously, thanks to Paladin's aura you're always in roll range but the class has no failure proofing tools so you're still failing those high end checks 50% of the time in your off-saves)

Not so good for:
- Fighter (though still good, it comes online much earlier when it counts while Indomitable takes infuriatingly long and is infuriatingly limited)
- Monk (though same as above, Diamond Soul comes really late and before then you're just another normal roller)

And of course, this is just considering saves: the less "save stress" you have on the feat, the more you can afford to use it on enemy crits, Initiative, skill checks, etc. Fighter and Monk are both very interested in what amounts to 3/day adamantine armor++ (make enemy reroll a critical hit making it about a 50% miss, 45% normal hit, 5% crit instead of 100% crit) and anyone with strong actions (this can be Action Surging Fighter removing a key enemy but more commonly someone with spellcasting dropping AOE CC to make enemy unable to kill you with alpha strike) is really, really interested in the Initiative rerolls.

[Dork_Forge]

It's a good feat, if you only have one combat a day and not much else then it can be a pretty big shift to outcomes, but as long as you're not that kind of table it's mostly just boring and eh.

One of my current groups has a Chronurgy Wizard with Lucky, he is since no longer in the group and there hasn't really been any big changes, nor did he really offset things much with that build (and he used both chronal shifts and all luck points every LR).

[Contrast]

...there's something fundamentally different about seeing an unfavorable roll that would potentially lead to disaster get rerolled, over and over again.

I will say my experience of seeing Lucky being used has been very different.

Lets say that generally speaking you'll be using Lucky to reroll a bad roll. Sometimes though you'll be rerolling a middling or even high roll just because you really don't want to fail. Its at its best to reroll rolls you have a high chance of success on but easy rolls are less likely to be critical rolls (plus you're obviously more like to pass them in the first place).

You get 3 Lucky uses per day. If we squint really hard and pretend this is proper statistics, we can assume that of those rolls we'll get one low one, one middling one and one high one. The low one probably isn't going to help us. The middling one might help us, might not and the high one probably will unless we're just pushing our luck on something difficult.

What this means is that in practice Lucky isn't a get out of jail free card 3 times a day. Factoring in that people will often want to save a Lucky point in case something happens which might mean they sit on it and end up burning it for something unimportant or finishing the day with one unspent and its actually more like a get out of jail free card probably once a day.

How good that is depends on the length of your days of course but my experience of watching other people use Lucky has definitely been that the most common use of a Lucky die is 'I'll use Lucky! ...still failed'.

[aguaracu]

Lucky can be useful in combat and fantastic on non-combat days. A day in town might involve a few important persuasion or investigation checks for example, you can effectively get advantage on the really useful rolls. The versatility of Lucky is great for versatile characters, but the player needs to be tactically astute (without being too meta-gamey).

Also, I feel that combat in D&D can be much less risky than skill checks. If the fight is between two big sacks of hitpoints then some failures are to be expected and it's the average DPR that matters[1]. It is, on the other hand, quite easy to run into crucial skill checks: fail that Stealth check or make a terrible Persuasion roll and what could have been a simple job becomes horribly complicated. The swinginess of the D20 system means that this can happen even when you have high abilities in the skills required. Needing to chain several checks together can make using social engineering or stealth a losing proposition.

For example, consider a heist where a stealthy infiltration needs:
1 Deception check (to enter the castle disguised as a chambermaid).
1 Athletics check (to climb in a window).
1 Stealth check (someone walks underneath as you hang from the windowsill).
1 Investigation check (to find a trap that will raise the alarm).
1 Thief's Tools check (to disarm said trap).
Failure in other checks (such as picking locks) may only result in small delay.

Assume in this scenario that your character has a 90% chance of succeeding in each check (you are good at this). With the raw 90% chance of success for each roll:
0.9**5 = 0.59
So you have a 59% chance of success. There's a 41% chance that you will have to kill everyone in the castle :-O

That's only just better that even odds so lets try some ASIs to improve your chances.

Two of the skill rolls are Dex so let's try an ASI to give +2 to Dex:
0.9**3 * 0.95**2 = 0.66
66% chance of success.

Two ASIs for +4 to Dex (making the Dex checks auto-succeed):
0.9**3 * 1.0**2 = 0.73
73% chance of success.

Two ASIs in different stats:
0.9**2 * 0.95**3 = 0.69
69% chance of success.

Three ASIs in different stats:
0.9 * 0.95**4 = 0.73
73% chance of success.

Four ASIs in different stats:
0.95**5 = 0.77
77% chance of success.

For Lucky, assuming you automaticaly re-roll every failure, I get a probability of success of 95%.
We group the 2^5 = 32 scenarios of success or failure in each roll into 6 categories: those where there are from 0 to 5 failures in the initial roll. For example where there is one failure there are 5 choose 1 = 5!/(1! * 4!) =5 possible ways this could happen (corresponding to the 5 scenarios where the 1st roll fails, or the second etc.).

# Raw failures 5 choose n Raw success Raw failure Lucky success Lucky failure
0 1 0.9**5 = 0.590 0.590
1 5 5 * 0.9**4 * 0.1**1 = 0.328 5 * 0.9**4 * 0.1 * 0.9 = 0.295 0.033
2 10 10 * 0.9**3 * 0.1**2 = 0.073 10 * 0.9**3 * 0.1**2 * 0.9**2 = 0.059 0.014
3 10 10 * 0.9**2 * 0.1**3 = 0.008 10 * 0.9**2 * 0.1**3 * 0.9**3 = 0.006 0.002
4 5 5 * 0.9**1 * 0.1**4 = 0.0005 0.0005
5 1 0.1**5 = 0.00001 0.00001
Total probability 0.590 0.410 0.951 0.049
This assumes that you know exactly what the DC for each ability check is and so this is too optimistic. I don't, however, think that this will wipe out the difference between this and the other options. If you look at the table in the spoiler above you will see that scenarios where you fail multiple raw rolls are quite rare and make little difference to the overall probability. For example, if you had two highish rolls and three lowish rolls then a policy of re-rolling all the lowish rolls would cover the one or two that would otherwise be failures without running out of Luck.
At level 11 Rogues do get Reliable Talent, which makes Rogues the best at what they do in tier 3. Everyone else, including Bards and low-level Rogues, have this problem.

Some DMs will - formally or informally - use a policy of Success with Consequences: if you narrowly fail a check they will give you a success but with extra complications. Eg. the guard lets you in but helpfully tells someone to mentor the new chambermaid and show you the ropes. Done well this can make using a range of skills viable and fun.
If the DM won't (or can't, not everyone is good at everything) do this well then trying to use skills to accomplish things can feel discouraging and doomed to failure. This is a question of DM and player creativity (and being on the same wavelength), but Lucky can help.

Apologies for the length, this kind of ran away from. All IMHO.

[1] Combat involves more than just knocking hitpoints off each other of course but it's fairly widely recognised that straight "save or die" combat scenarios are not fun to play. I think that the same is true of non-combat scenarios but that this is not so widely appreciated.
As well as the defensive points mentioned by others you can deliberately make certain in-combat attacks or skill checks very bad for the enemy when you succeed. Eg. landing a hit when casting Contagion or making checks for initiative, grapples, Telekinesis or Counterspell.

Edit: 2nd paragraph, last sentence re-worded. Stray word removed from the last paragraph.