# D&D 3E/3.5Analyzing 4 Methods for Converting 3.5 and Pathfinder to 5E

#### Carcharoth

##### Explorer
The article on EN World titled Attempt at Conversion doc to convert 3.5 edition and Pathfinder monsters to D&D Next - On the Fly is an excellent primer in quick monster conversion for DnD Next, which although written before the release of 5e, is still relevant. Member jamesmanhattan breaks down the reasoning behind his method, provides insight into the relationships of four primary stats (AC, HP, AB, and DPR) to CR, and offers additional informative afterthoughts and clarifications. Starting at comment #17, KnightCa further adds to the conversation by providing a few noteworthy adjustments.

After messing around with various monster conversion techniques, and the Monster Statistics by Challenge Rating guide of the DMG on page 274, I decided to apply and compare three different methods of conversion to the Pathfinder Database available on d20pfsrd. Not knowing enough about third-party monster creation procedures, I limited the sample to monsters from Pathfinder Bestiaries 1 through 4 to insure a consistent base for comparisons. When the recent Conversions to Fifth Edition doc from Wizards was released, I incorporated it as well, and the end results were very pleasing. In the end, I figured I’d share the results in the off chance that others may find the information useful.

Note: while gathering the sample, I inadvertently deleted a few creatures. Additionally, the spreadsheet contained several duplicate monsters (by name) with varied statistics, so they are included as well. Although not every monster from Bestiaries 1 to 4 is included, a sample of 1413 were compared. Once the spreadsheet is cleaned up a bit, I intend to post it for review.

### Summary​

Presented first is the end result of an amalgam of three existing conversion methods with additional refinements, followed by the details of establishing a baseline, and the analysis, comparison, and contrast of each method.

Method
 Armor Class: (AC + Touch AC)/ 2 rounded down If CR < 21 and AC > 20, Then AC = 20. If CR > 20 and AC > 20, Then (AC-20)/ 4 + 20 rounded down. Hit Points: As is. Proficiency Bonus: (CR - 1)/ 4 + 2 rounded down. Ability Modifiers: If AM > 5, then (AM - 5)/ 3 + 5 rounded down. Attack Bonus: Appropriate Ability Modifier + Proficiency Bonus. Damage: As is. Save DC: 8 + Appropriate Ability Modifier + Proficiency Bonus. Saving Throws: Appropriate Ability Modifier, + Proficiency Bonus if Good Save.

Coupling the conversion doc from Wizards with the compression of higher Ability Mods, and addition of Proficiency Bonuses to calculate remaining attributes, provides the most overall consistency with 5e. The added adjustments do, however, make this version a little slower to process “on the fly”, but with a little practice, and a cheat-sheet, you can do the calculations pretty quickly in your head.

#### Details​

Following are five methods for converting 3.5 and Pathfinder monsters to 5e. Each contains a brief analyses, and a few inexpert opinions. Please keep in mind, like jamesmanhattan’s article, the objective of each method is to convert monster’s stats to make them 5e-ish, not to match CRs or make each monster identical to its 5e counterpart.
1. Pathfinder Stats Converted by the Book
2. jamesmanhattan’ Method
3. KnightCa’s Method
4. Wizards’ Method

### Pathfinder Stats Converted By the Book: DMG pg. 274​

Method
1. Find the CR for its corresponding HP, and adjust the CR up or down by one for every two points of difference between the original AC and the corresponding AC for this CR. This is the Defensive CR.
2. Find the CR for its corresponding DPR, and adjust the CR up or down by one for every two points of difference between the original Attack Bonus and the corresponding Attack Bonus for this CR. This is the Offensive CR.
3. The final CR is the average of the Defensive and Offensive CRs rounded up.

Results
Note: The reason the CRs are grouped in each method is to measure its effects in relation to the Monster Statistics table, which can then be compared and contrasted to the other methods.
 Distinct​ Total​ %​ -10​ 1​ 0%​ -9​ 1​ 0%​ -7​ 1​ 0%​ -6​ 4​ 0%​ -5​ 30​ 2%​ -4​ 70​ 5%​ -3​ 169​ 12%​ -2​ 242​ 17%​ -1​ 290​ 21%​ 0​ 192​ 14%​ 1​ 98​ 7%​ 2​ 64​ 5%​ 3​ 38​ 3%​ 4​ 39​ 3%​ 5​ 34​ 2%​ 6​ 33​ 2%​ 7​ 20​ 1%​ 8​ 21​ 1%​ 9​ 21​ 1%​ 10​ 21​ 1%​ 11​ 14​ 1%​ 12​ 9​ 1%​ 13​ 1​ 0%​ 0.06​ 1413​ 100%​ +/- 2​ 991​ 70%​ +/- 3​ 1125​ 80%​
70% of CRs fall within +/- 2 of CR -1.
• The group of converted CRs have a weighted average of 0.06, which is surprisingly close, but it’s distributed over a fairly large span of 23 CRs.
• Although CRs fall quickly in the negative range, they level out over a large area of CRs in the upper range, so the bell’s peek is actually around -1 instead of 0.
Not too shabby, but the distribution is a bit messy, and the sheer volume of extremely high AC and Attack Bonus numbers make it hard to get a sense of how they fit into the 5e world. These numbers do, however, give us a baseline with which to compare various conversion methods.​
The conversion methods to follow will tighten the curve, and bring stats closer in line with 5e.​

### jamesmanhattan’s Conversion Guidelines​

Method
 AC: AC - CR. Attack Bonus: (CR/ 2) + 3, includes all multiple attacks. Damage: As is. Save DC: (DC - 10)/ 2 + 10. Hit Points: As is. Saving Throws: Save Bonus/ 2 (leave negatives as is). Ability Checks: Save Bonus/ 2 (leave negatives as is). Skills: Save Bonus/ 2 [+ 5 if trained]. Initiative Bonus: As is.

Results
 Distinct​ Total​ %​ -14​ 1​ 0%​ -12​ 2​ 0%​ -10​ 4​ 0%​ -9​ 15​ 1%​ -8​ 23​ 2%​ -7​ 49​ 3%​ -6​ 132​ 9%​ -5​ 291​ 21%​ -4​ 342​ 24%​ -3​ 279​ 20%​ -2​ 158​ 11%​ -1​ 66​ 5%​ 0​ 30​ 2%​ 1​ 12​ 1%​ 2​ 6​ 0%​ 3​ 2​ 0%​ 4​ 1​ 0%​ -3.921​ 1413​ 100%​ +/- 2​ 1202​ 85%​ +/- 3​ 1317​ 93%​
85% of CRs fall within +/- 2 of the weighted average.
• The weighted average dropped significantly down to -3.9.
• Since HP and DPR remain the same, the relationship between HP and DPR to AC and AB become quite apparent.
• The AC and AB numbers look much better.
• The CR range shrank from 21 to 17, and the curve looks much better.
The adjustments make a significant difference, and the overall stats make everything look more 5e-ish.​

Method
 AC: AC - CR (2 minimum) + 2 Proficiency Bonus: (CR - 1)/ 4 + 2 Ability Saves: Ability modifier Proficient Skills: Ability Modifier + Proficiency Bonus Attack Bonus: Highest of STR/ DEX Modifier + Proficiency Bonus HP: Large Size (higher than CR 2): Increase HP by 50%. Did not apply this step.

Results
 Distinct​ Total​ %​ -13​ 1​ 0%​ -12​ 1​ 0%​ -11​ 1​ 0%​ -10​ 1​ 0%​ -9​ 3​ 0%​ -8​ 5​ 0%​ -7​ 12​ 1%​ -6​ 33​ 2%​ -5​ 99​ 7%​ -4​ 199​ 14%​ -3​ 335​ 24%​ -2​ 368​ 26%​ -1​ 158​ 11%​ 0​ 81​ 6%​ 1​ 33​ 2%​ 2​ 25​ 2%​ 3​ 32​ 2%​ 4​ 18​ 1%​ 5​ 6​ 0%​ 6​ 1​ 0%​ 7​ 1​ 0%​ -2.329​ 1413​ 100%​ +/- 2​ 1141​ 81%​ +/- 3​ 1273​ 90%​
81% of the CRs fall within +/- 2 of the weighted average.
• The weighted average is -2.3, so it brought the CR up by ~1.5x due to the combination of higher AC and AB.
• The curve flattened a bit, which accounts for the drop to 81% within +/- 2 CR. Still pretty darn good.
• Incorporates a Proficiency Bonus, which is nice to have, and happens to be spot on with the table.
• Raises the AC which helps, but increases the number of excessively high ACs from the previous method.
• Uses Ability Modifiers, which is another step in the right direction, but ~16% of monsters have STR or DEX scores greater than 30, which skews the results significantly.
• AB now relies on Ability Mods, which is better, but mods are too high.
• The range of CR variance grew to 21; although, only a total of 24 monsters fall within the lowest 6 CRs, so that’s not significant.
A slight drop in numbers from jamesmanhattan’s method, but the addition of PB and Ability Mods is very useful and more consistent with 5e.​

### Wizards Conversions to fifth Edition​

Method
 Armor Class: (AC + Touch AC)/ 2 rounded down. Max 22. Hit Points: As is. Attack Bonus: Appropriate Ability Modifier + 3. Damage: As is. Save DC: Appropriate Ability Modifier + 10. Saving Throws: Appropriate Ability Modifier, + 3 if Good Save.

Results
 Distinct​ Total​ %​ -11​ 1​ 0%​ -10​ 2​ 0%​ -8​ 4​ 0%​ -7​ 8​ 1%​ -6​ 16​ 1%​ -5​ 70​ 5%​ -4​ 179​ 13%​ -3​ 350​ 25%​ -2​ 388​ 27%​ -1​ 211​ 15%​ 0​ 77​ 5%​ 1​ 37​ 3%​ 2​ 35​ 2%​ 3​ 25​ 2%​ 4​ 8​ 1%​ 5​ 1​ 0%​ 6​ 1​ 0%​ -2.195​ 1413​ 100%​ +/- 2​ 1205​ 85%​ +/- 3​ 1312​ 93%​
85% fall within +/- 2 of the weighted average.
• The weighted average is -2.2.
• The ACs look the best of all methods so far, and are capped at 22 to bring them in line with 5e.
• Uses Ability Mods, which is good, but are still too high to fit into 5e.
• The calculations used for AB, Save DC, and Saving Throws are pretty nice in that they imply Ability Scores here are really a combination of 5e-ish sized mods and PB if they were combined.
• The CR range is 16.
Although the percentages match jamesmanhattan’s method, the better ACs and use of Ability Mods for AB, Save DC, and Saving Throws make this method the most consistent with 5e.​
Unfortunately, even though it’s my favorite method so far, I still dislike the excessively high Ability Mods, and the lack of PB -- so I made two more adjustments.​

### Wizards Conversion + Three Adjustments​

Method
 Armor Class: (AC + Touch AC)/ 2 rounded down If CR < 21 and AC > 20, Then AC = 20. If CR > 20 and AC > 20, Then (AC-20)/ 4 + 20 rounded down. Hit Points: As is. Proficiency Bonus: (CR - 1)/ 4 + 2 rounded down. Ability Modifiers: If AM > 5, then (AM - 5)/ 3 + 5 rounded down. Attack Bonus: Appropriate Ability Modifier + Proficiency Bonus. Damage: As is. Save DC: 8 + Appropriate Ability Modifier + Proficiency Bonus. Saving Throws: Appropriate Ability Modifier, + Proficiency Bonus if Good Save.

Results
 Distinct​ Total​ %​ -11​ 2​ 0%​ -10​ 1​ 0%​ -8​ 3​ 0%​ -7​ 13​ 1%​ -6​ 28​ 2%​ -5​ 77​ 5%​ -4​ 250​ 18%​ -3​ 415​ 29%​ -2​ 326​ 23%​ -1​ 159​ 11%​ 0​ 67​ 5%​ 1​ 39​ 3%​ 2​ 24​ 2%​ 3​ 7​ 0%​ 4​ 2​ 0%​ 5​ 0​ 0%​ -2.576​ 1413​ 100%​ +/- 2​ 1217​ 86%​ +/- 3​ 1333​ 94%​
86% fall within +/- 2 of the weighted average.
• The weighted average is -2.6.
• AC are computed using the same method as Wizards accept they are capped at 20 for CRs 20 and under, and compressed for CRs over 20, which will max out at 25.
• Proficiency Bonus is computed for further calculations, and matches 5e progression.
• Ability Mods are brought in line by compressing the top tier.
• AB, Save DC, and Saving Throws are now calculated “by the book” because we can use PB.
• The CR range is 16.
The final adjustments made to the Wizards method make this conversion process the most consistent with 5e. The compression of Ability Scores and inclusion of PB allow AB, Save DC, and Saving Throws to be calculated “by the book”, while the new AC caps and compression wrap up the whole process pretty neatly.​

### Notes on Compression and Maximums​

#### AC ​

After analyzing the 5e Monster Manual, it appears that seven monsters of CR 20 and below have an AC 20, while only seven monsters over CR 20 have an AC greater than 20. It seems pretty safe (for now) to presume this a general rule, so I chose to incorporate it until the future shows otherwise.

Of 1413 monsters sampled, 331 have an AC greater than 25, while 32 are over CR 20 and have an AC greater than 20 -- max 50.

The average of AC and Touch AC is a great touch, and brings the maximum AC down from 50 to 40. That being said, to make things tidy, I chose to compress the AC range of 20 to 40 into a range of 20 to 25. It seems that several creatures should be closer to a tarrasque-like AC of 25 instead of the 22 cap suggested by Wizards.

The main benefit for compressing ACs is that injects a little more consistency to the game. You can just as easily not compress them and, at worst, it will have only a negligible effect on the overall stats. I would however cap ACs to 20 on CR 20 and less.

#### Ability Scores​

Strength scores in Pathfinder Bestiaries 1 through 4 go as high as 56. About 55% of monsters have a STR or DEX above 20, and about 16% have a STR or DEX greater than 30. The sheer number of excessively high scores skew the results significantly when not adjusted.

The Ability Mod formula presented above compresses the bonuses of existing Ability Scores 20 and higher from a range of 5 to 20 down to a range of 5 to 10. There is one exception: The Great Old One is the only creature that has an original STR bonus of 23, which results is a 5e bonus of +11. It is left to the DM’s discretion whether or not to cap bonuses at +10, and related Ability Scores to 30.

Since both 3.5 and Pathfinder stat blocks provide Ability Scores rather than Ability Mods, you will need to calculate the new bonuses yourself. Luckily it’s easy, and can be done one of two ways; derive the new bonuses from existing Ability Scores, or derive them from newly calculated scores. The following formulas support these methods.
1. The standard formula for calculating Ability Mods is (Ability Score - 10)/2 rounded down.
2. The formula for calculating new Ability Mods for Ability Scores 20 and over is (Ability Score - 20)/ 6, rounded down, + 5.
3. The formula for calculating new Ability Scores 20 and over is (Ability Score - 20)/3, rounded down, + 20.
Following are a few examples to illustrate usage of the formulas above.
1. If a BBEG has an original STR of 19 (any score under 20), then the Strength Bonus is (19-10)/2, rounded down, which is +4.
2. If a BBEG has an original STR of 37 (any score 20 and over), then the new Strength Bonus is (37-20)/6, rounded down, plus 5, which is +7.
3. The new STR of the BBEG in example 2 above is (37-20)/3, rounded down, plus 20, which is 25.
Note: If you calculated a new score first (step 3), you can use either the formula from step 2 on the original score, or use step 1 on the new score. Both will result in +7.

### Conclusion​

Coupling the conversion doc from Wizards with the compression of higher Ability Mods, and addition of Proficiency Bonuses to calculate remaining attributes, provides the most overall consistency with 5e. The added adjustments do, however, make this version a little slower to process “on the fly”, but with a little practice, and a cheat-sheet, you can do the calculations pretty quickly in your head.

As alluded to previously, there is a lower HP and DPR average for 3.5 and Pathfinder monsters. Overall, the average monster has 85% fewer HPs, and inflicts 77% less DPR for the expected 5e CR. The lower HP and DPR in relation to the modified AC and AB values account for the lower weighted averages in the methods tested.

The results are not perfect, and do not account for relevant abilities or traits that increase Effective AC, HP, CR, etc. But that’s OK. If you recalculate 5e monsters using the table in the DMG, you will find discrepancies there as well. As the DMG states, the art of monster making is more than just number crunching. Just add flavor as desired.

The objective is to make stats more usable in 5e, not to bring CRs between the two systems in line with each other. As such, when converting monsters, leave the original CR as is. The adventures they are used in are balanced around that CR, not a re-calculated one.

Consider the following: the Pathfinder Rust Monster is CR 3, while the 5e Rust Monster is CR ½. After converting the Pathfinder version, leave it as CR 3, and make other adjustments as you see fit. You now have the added benefit of a more worthy adversary for adventurers of higher levels. If you use them this way, you end up with thousands of variations, allowing you to easily add diversity to your campaign. A win-win, especially with all the readily available OGL content at your disposal.

If you find errors, or have any suggestions, please let me know.

#### Attachments

• dd-feature-logo.jpg
48.1 KB · Views: 4,008
Last edited:

#### Morrus

##### Well, that was fun
Staff member
That's an enormous amount of work - thank you! I'm bookmarking this for a thorough read later!

#### Barachiel

##### First Post
Step One for converting 3.5 to 5e

DON'T.

It was never meant to translate. Converting your Paladin/Cavalier/Divine Champion is not going to net you the same class features in 5E that you thoroughly enjoyed in 3.5e. Simple as that. That goes for any class.

#### Waller

##### Hero
Step One for converting 3.5 to 5e

DON'T.

It was never meant to translate. Converting your Paladin/Cavalier/Divine Champion is not going to net you the same class features in 5E that you thoroughly enjoyed in 3.5e. Simple as that. That goes for any class.

Didn't read the article, huh? I get it. Reading is hard stuff.

#### Morrus

##### Well, that was fun
Staff member
Step One for converting 3.5 to 5e

DON'T.

It was never meant to translate. Converting your Paladin/Cavalier/Divine Champion is not going to net you the same class features in 5E that you thoroughly enjoyed in 3.5e. Simple as that. That goes for any class.

The article is about monster conversion. There's an excellent clue in the first sentence.

#### dd.stevenson

##### Super KY
Yeah, this article matches my own experiences for the most part. One thing I have found with pathfinder is that it never hurts to add on about 15% hp to match the challenge weightiness the encounters should have had in the original adventure.

#### Dungeonosophy

##### Legend
I appreciate your nuanced scientific offering. Thanks!

#### Carcharoth

##### Explorer
Yeah, this article matches my own experiences for the most part. One thing I have found with pathfinder is that it never hurts to add on about 15% hp to match the challenge weightiness the encounters should have had in the original adventure.

I originally did the same thing but I’m still working on finding a more consistent method. There is a fairly tight progression when it comes to HP and CR in relation to the expected 5e values.

For instance, the lower CR ranges are closer to 20% of the expected 5e values, while the higher ranges are closer to 100%. Not until CR 2 does a creature reach ~50% of the expected HP, CR 7 to reach 71%, or CR 15 to reach 98%. It appears that one should grant a higher percentage of HP to creatures with lower CRs, while granting a lower percentage to those with higher CRs.

View attachment 72211

DPR on the other hand is pretty much all over the place. I fully expect the DPR to vary more since many creatures use magic vs. standard melee, but the DPR relationship is definitely far more variable.

View attachment 72212

#### dd.stevenson

##### Super KY
I originally did the same thing but I’m still working on finding a more consistent method. There is a fairly tight progression when it comes to HP and CR in relation to the expected 5e values.

For instance, the lower CR ranges are closer to 20% of the expected 5e values, while the higher ranges are closer to 100%. Not until CR 2 does a creature reach ~50% of the expected HP, CR 7 to reach 71%, or CR 15 to reach 98%. It appears that one should grant a higher percentage of HP to creatures with lower CRs, while granting a lower percentage to those with higher CRs.

Have you looked at something like CR * Con Mod?

I'm not gonna lie; I just use a regex macro in word to rewrite the important stats, so your more complex percentage change by level method works just swell for me

#### dwayne

I use my own conversion chart, my imagination and adjust as need be to match the flavor and the intent of the creature I want to use. In other words I make up the stats that I feel best suit the creature and what it was meant to be as per the original creatures stats. But the charts are nice for the ones who just need the facts and a ruler to draw the lines, was a good read

#### neobolts

##### Explorer
Excellent work, although my eyes glazed over a little bit when I hit the charts.

It looks like a very logical and simple approach to the reconciling 5e bounded values with the 3.5's boundless values. I take it the Summary method at the top is the "Carcharoth method" you refined from the results of analyzing the other 5 methods?

#### Wrathamon

I take it the Summary method at the top is the "Carcharoth method" you refined from the results of analyzing the other 5 methods?

That was my take as well

#### Carcharoth

##### Explorer
Have you looked at something like CR * Con Mod?

I'm not gonna lie; I just use a regex macro in word to rewrite the important stats, so your more complex percentage change by level method works just swell for me

Unfortunately, because the lower CRs typically have lower Con Mods (smaller boost), while higher CRs typically have higher Con Mods (larger boost), it doesn't really match the type of progression. The lower CRs really need a higher boost in relation to higher CRs.

BTW, regex and excel are friends of mine

#### Carcharoth

##### Explorer
Excellent work, although my eyes glazed over a little bit when I hit the charts.

It looks like a very logical and simple approach to the reconciling 5e bounded values with the 3.5's boundless values. I take it the Summary method at the top is the "Carcharoth method" you refined from the results of analyzing the other 5 methods?

After reading the Wizards doc, I did make a conscious effort to bind the values. Compression seemed to be the best bet, as it allowed me to start using basic 5e rules to calculate AB, Save DCs, etc.

There was a request earlier to provide a summary at the beginning of the post because it is so long, so I provided the end results there. It appears I can refine that area -- I there something I should say to make it a bit clearer?

On the subject of charts, I talked to someone about the numbered charts... would it be better/ more helpful to provide graphs with a summary rather than the detail? I can whip them up pretty quickly if it would improve the explanation.

#### neobolts

##### Explorer
After reading the Wizards doc, I did make a conscious effort to bind the values. Compression seemed to be the best bet, as it allowed me to start using basic 5e rules to calculate AB, Save DCs, etc.

There was a request earlier to provide a summary at the beginning of the post because it is so long, so I provided the end results there. It appears I can refine that area -- I there something I should say to make it a bit clearer?

On the subject of charts, I talked to someone about the numbered charts... would it be better/ more helpful to provide graphs with a summary rather than the detail? I can whip them up pretty quickly if it would improve the explanation.

I think one sentence at the start of the summary would improve the readability of the whole post, i.e. "First let me show you a composite formula I made from the best performing aspects of five different methods, and then support my conclusions with a detailed breakdown each of those five methods."

#### Carcharoth

##### Explorer
I think one sentence at the start of the summary would improve the readability of the whole post, i.e. "First let me show you a composite formula I made from the best performing aspects of five different methods, and then support my conclusions with a detailed breakdown each of those five methods."

Thanks for the excellent suggestion. I updated the post; however, I don't see the changes yet, even though they appear in the editor. Maybe it will show up a little later.

Regardless, yesterday, I came across the following tweet from Jeremy Crawford regarding monster AC on http://www.sageadvice.eu/2015/11/30/no-monster-has-ac-above-25/. All though I presume most people anticipated this, its still nice to hear an officially unofficial answer.

View attachment 72244

Replies
0
Views
772
Replies
0
Views
931
Replies
8
Views
1K
Replies
0
Views
218
Replies
4
Views
681