Persist Rods?

[killianh]

I'm wondering if it would be too broken to allow Rods of metamagic: Persist?

The prices I came up with were 50k for the lesser, 119k for the normal, and 269k for the greater versions respectively. Considering that it's pretty easy to DMM: persist Arcane spells as it stands and there are a number of ways to do so (cleric dip + alternative spell source for example) I'm wondering if this would be a game breaking item?

Thoughts?

[Hirax]

I'm very skeptical of granting new ways to access one of the most powerful things in the game. I wouldn't say never, but I'm having a hard time visualizing when I'd allow it.

[Tvtyrant]

See, at 269 K the only class able to easily afford it would be the Artificer. Anything that benefits only tier 1s is something that should be heavily scrutinized.

[killianh]

See, at 269 K the only class able to easily afford it would be the Artificer. Anything that benefits only tier 1s is something that should be heavily scrutinized.

Or purchase gets delayed until epic levels. What are your thoughts on the lower versions?

[Tvtyrant]

Or purchase gets delayed until epic levels. What are your thoughts on the lower versions?

Well the least is fine IMO, since you would only be able to do with it what you would with unmodified persist. The normal one would be quite potent; I'm not sure if it is in the "must have" area or not.

[HunterOfJello]

The normal version price of around 118,000-119,000gp seems right, since it follows a linear progression.

I'm not sure why the lesser version doesn't follow a linear progression, but it doesn't. Using a power regression the price would be around 56000 (56728gp to be exact) and using a second order polynomial regression puts it around 91,000gp. The second regression is a better fit to the line, but an inappropriate price compared to the normal version. I would set the price for the Lesser Persist Rod at a minimum of 57,000gp (if it was allowed at all).

[killianh]

The normal version price of around 118,000-119,000gp seems right, since it follows a linear progression.

I'm not sure why the lesser version doesn't follow a linear progression, but it doesn't. Using a power regression the price would be around 56000 (56728gp to be exact) and using a second order polynomial regression puts it around 91,000gp. The second regression is a better fit to the line, but an inappropriate price compared to the normal version. I would set the price for the Lesser Persist Rod at a minimum of 57,000gp (if it was allowed at all).

I was following this from this post (http://www.enworld.org/forum/d-d-legacy-discussion/196460-meta-magic-rod-persistant-spell-cost.html)


Here is a formula I came up with a couple years back:

I came up with a simple solution, except there is one rod which doesn’t match up (the lesser quicken rod).

Here is the equation:

((L*2)-1)*(M^2)*300gp

L = the level adjustment of the metamagic feat which the rod duplicates
M = the maximum level of spell which the rod is able to effect

Code:
type M L cost Calc
lesser 3 1 3000 2700
regular 6 1 11000 10800
greater 9 1 24500 24300
lesser 3 2 9000 8100
regular 6 2 32500 32400
greater 9 2 73000 72900
lesser 3 3 14000 13500
regular 6 3 54000 54000
greater 9 3 121500 121500
lesser 3 4 35000 18900 (why?)
regular 6 4 75500 75600
greater 9 4 170000 170100

*estimation for persistent rods*
lesser 3 6 ----- 29700
regular 6 6 ----- 118800
greater 9 6 ----- 267300 (maybe x10 for epic?)
If anyone can derive a better equation, that would be great. I’d like to see if someone can generate an exact formula.

Rounded some of the numbers off and upped the lessor cost to match the kind of jump the lesser Quicken Rod had.

EDIT: I would love to know the math system(s) you used for that.

[HunterOfJello]

EDIT: I would love to know the math system(s) you used for that.

I didn't spend much time calculating anything. I mainly put the numbers into excel and used the Regression function in the Data Analysis addon. Most people don't learn how to do Regression Analysis, even in college, so it seems more impressive than it is. Then I used the Trendline function in the Layout section of the chart tools. This set up all the different lines of best fit.

~

There are several methods that are often used to find the line of best fit for any given set of numbers. They're usually used for large scatter plots in order to gain useful predictive ability out of them. The modeling method this is all involved in is called Regression Analysis.

Linear Regression (http://en.wikipedia.org/wiki/Linear_regression) finds a solid y=mx+b line that best fits for the data given. This can be very useful for all sorts of things, but is generally not the most accurate method of predicting data.

Power Regression finds a different line of best fit generally based on the equation y=b+x^m. This creates a line that fits well on data that resembles a logarithm.

Polynomial Regression is pretty cool and is based on a line of best fit based on a polynomial equation. A second order polynomial would have the general equation of y=a+bx+cx^2. A third order polynomial would have the general equation of y=a+bx+cx^2+dx^3. A fourth order would add in an x^4 and so on. The polynomial regression has the interesting property that as you continue to go to higher orders, you get a line that is increasingly closer to the data that you've input. Even if you put x=1,2,3,...,100 and y=random numbers between 1 and 100, then you can still get an amazingly close line of best fit if you increase the order high enough. It's pretty cool. Anyway, the polynomial regression often gives the closest line of best fit to what the data shows, but often runs off in extreme directions outside of the original data inputted for this sort of thing. That's why the equation came up with 91000gp or so instead of half that much.

~

You can access these lines of best fit in Excel if you put in the information into cells, highlight it all, and then set up a scatterplot chart. Click on the scatterplot chart and then click Layout, then Trendline (More Trendline Options). A window should pop up if you click More Trendline Options, and you can choose between the different lines of best fit and have them forecast predictions going forward or backward along the x-axis.

That's the method I used to find the numbers. I only used the numbers for metamagic rods that increased the spell level by 1, 2, 3, and 4 because the 0th level metamagic rods seem to have very odd methods of calculation. Their price is based around having a feat instead of not having a feat, not having a feat and increasing a spell's slot level. It also didn't calculate well and wasn't worth disturbing everything else for.

~

The equation he listed is likely closer in general form to what WotC used, but it is obviously far on in it's predictive capabilities for lesser rods seeing as how it predicted lesser +4 metamagic rods as being 18900 instead of 35000. WotC is fond of using methods that involve a constant number like 300 being multiplied by some sort of other function. They do that with scrolls and things.

The reason the lesser rod doesn't match up could be due to an input error on WotC's part that was never corrected. The regular and greater rod versions have much smoother lines in comparison to the lesser. The lesser versions could also have been modified so that they were more in line with what WotC thought they should be valued as.

[GenghisDon]

I'm wondering if it would be too broken to allow Rods of metamagic: Persist?

The prices I came up with were 50k for the lesser, 119k for the normal, and 269k for the greater versions respectively. Considering that it's pretty easy to DMM: persist Arcane spells as it stands and there are a number of ways to do so (cleric dip + alternative spell source for example) I'm wondering if this would be a game breaking item?

Thoughts?

If one's game is already broken with DMM & nightsticks then letting any spellcaster do the same for more $ is probably fair.

I'd have debuff's everywhere to discourage the tactic's abuse, myself

The last one is an epic level item, and probably deserves the skyrocking price such get.