Arkham Horror (3rd Ed) Statistics

By ricedwlit, in Arkham Horror

As noted in another thread I’ve decided to take on the challenge of collecting data on Arkham Horror (3rd ed) game plays. Of course the work is not fully my own since I cribbed liberally from the work Tibs did for Arkham Horror (2nd ed) and Eldritch Horror.

You can submit you play session information here

You can view submissions and derived statistics here .

Note: tabs that contain spoilers include "spoilers" in the name (with one exception, namely the tab containing the submissions - but then it should be obvious that it will contain spoilers. For those of you curious as to what was submitted who want to avoid spoilers there is a "Spoiler Free Submissions" tab).

If you have any questions / concerns please post here or contact me at [email protected]

Edited by ricedwlit
Include link for statistics.

Thanks all for submissions to date (as of this post, 188 games)!

Some new stats have been added to the Scenarios tab (based on data from the info already present in the various spoilers tabs) regarding how often each scenario proceeds to the point where it become markedly different to win (and what the % loss if that point is reached)*. This roughly corresponds to the "Ancient One Awakens" or "Final Mystery" in AH (2nd ed) and Eldritch Horror respectively).

*: Well, this isn't true for all four scenarios - but to be consistent I didn't want to mark any scenario as special.

Edited by ricedwlit
Spelling error, clarification.

Hi. Tomorrow the game in Spain goes on sale.

When we play the games we will use this statistics system.

Thank you.

I'm guessing people are mostly using Tommy for the harder scenarios, because no way should he be showing up on the bottom 3 win rates.

I'm more astonished by the win vs loss ratio. So far I've lost 6 times out of 24 plays, but the first three defeats were tied to when the game was in a very early stage and was a lot harder to beat. After it reached a more or less stabilized form, I'm 18-3. And I've never lost a game with Tommy.

Edited by Julia
On 4/7/2019 at 2:04 PM, Julia said:

I'm more astonished by the win vs loss ratio. So far I've lost 6 times out of 24 plays, but the first three defeats were tied to when the game was in a very early stage and was a lot harder to beat. After it reached a more or less stabilized form, I'm 18-3. And I've never lost a game with Tommy.

That you have such a high win rate doesn't surprise me. :)

I suspect that you are an outlier regarding "grokking" the game and seeing what needs to be done to win. I know that for my part it took a while to completely grasp what to focus on in AH 3rd edition and my personal stats reflect this: 13 games with 6 victories, distributed across all scenarios (plus rotating through all investigators). Now that I've played each scenario multiple times I have a better sense on what to do to win. Coupled with selecting best investigators I'm sure my win rate will rise to above 75% (and for the Unmordoth it will likely hit 95% or more).

However, right now stats are dominated by people discovering them and perhaps recording their first game. And Approach to Azathoth is not easy - if you lose control of doom and let things escalate it will be hard to win (even more so if you had just spent clues to take a certain action).

That's very true (re: people just discovering the game might bias the actual difficulty of the game itself). And yes, Azzie is the most difficult from the core set; I remember playing it again a few months ago, when we got our retail copy, after some 10 months we ended the alpha, and we had a hard time trying to contain doom (we botched the reading of the scenario, and were able to steer it back on track only by a miracle).

Azzie's problem is the starting position for investigators: you can't cover properly the board and mantain doom under control in the first round, so you can easily have an anomaly on Mythos 1.

9 hours ago, Julia said:

Azzie's problem is the starting position for investigators: you can't cover properly the board and mantain doom under control in the first round, so you can easily have an anomaly on Mythos 1.

Yup, happened to my in my second play through with Azathoth. 😡

On a separate note: There are now over 350 entries (not bad for a game with no expansions yet).

For those who looking to accomplish a first: we've yet to have a win recorded for a six person game!

Weird that Tommy is near the bottom, while...Daniela is at the very top???? o_O

We've always considered Daniela to generally be a flat out weaker version of Tommy.

Good news : I will be updating the submission form to include the Dead of Night is expansion after I get it (it's should arrive in a day or two I'm told).

Bad news : I am going to want to play the game at least a few times before spoiling the resolutions for myself. This means it will likely be a few weeks (what with other responsibilities) before the form is updated. So for those of you playing it in the meantime who may want to enter their data later, you should know what will be needed, e.g
names of investigator(s), scenario played, outcome (victory or defeat) and how (in case of victory, likely just need to track the codex number).

Beware of the outcomes, it's not sure they are boolean... (talking in general, not necessarily of this expansion)

The form has just been updated for the Dead of Night expansion. I took the plunge to collect data for the scenarios even though I've yet to play them

The summary stats spreadsheet based on submissions has not yet been updated to reflect any data for the new games and/or investigators. It's just recording submissions for now so a few tables not dependent on scenario or investigators will update, but if you are curious as to how a given new investigator is doing you will have to wait until there is sufficient data to be meaningful.

On 10/4/2019 at 5:05 PM, Julia said:

Beware of the outcomes, it's not sure they are boolean... (talking in general, not necessarily of this expansion)

There were indeed multiple outcomes for one, however they could be loosely classified into victory/defeat. I'm sure at some point we'll get something else (e. "the world is reset ... you neither win or lose").

Edited by ricedwlit
Additional details

Update: Over the past few months data has been updated on the sheet. You can now see stats for the new investigators and scenarios from Dead of Night (including some tabs with detailed information on outcomes, as with prior ones these are labelled with "spoiler").

Fellow investigators!
I have taken over the task of maintaining and updating that stats form and sheet, and will try to do justice to Richard's fine work up until now.

Today I have implemented the following changes to the form:

  • Added the 4 scenarios and 8 investigators from Under Dark Waves
  • Added the 4 variant Recursive Echoes investigators, denoted by (RE)
  • Added the Arkham Nights variant Ashcan Pete investigator, denoted by (AN)
  • Added optional entry of game mode
  • Added optional entry of inevitable game number

And the following changes to the sheet:

  • Included the new scenarios in the basic scenario summary stats, i.e. times played, overall difficulty
  • Included all new investigators in the stats calculations
  • Renamed the existing Pete entries to Pete (AN), to reflect that they all were played using the variant version

The specific scenario stats sheets with spoilers for the 4 new scenarios will follow in the next update, and after sufficient data has been submitted for the data to be meaningful.

Please do no hesitate to send questions, comments or corrections to [email protected].

On 10/29/2020 at 3:22 PM, pjelstrup said:

Fellow investigators!
I have taken over the task of maintaining and updating that stats form and sheet, and will try to do justice to Richard's fine work up until now.

Today I have implemented the following changes to the form:

  • Added the 4 scenarios and 8 investigators from Under Dark Waves
  • Added the 4 variant Recursive Echoes investigators, denoted by (RE)
  • Added the Arkham Nights variant Ashcan Pete investigator, denoted by (AN)
  • Added optional entry of game mode
  • Added optional entry of inevitable game number

And the following changes to the sheet:

  • Included the new scenarios in the basic scenario summary stats, i.e. times played, overall difficulty
  • Included all new investigators in the stats calculations
  • Renamed the existing Pete entries to Pete (AN), to reflect that they all were played using the variant version

The specific scenario stats sheets with spoilers for the 4 new scenarios will follow in the next update, and after sufficient data has been submitted for the data to be meaningful.

Please do no hesitate to send questions, comments or corrections to [email protected].


TYVM! Any way we could take a quick look at scenario wins/losses by investigator count?

19 hours ago, Einlanzer80 said:


TYVM! Any way we could take a quick look at scenario wins/losses by investigator count?

That is an excellent idea! I'll try to incorporate that in the next update.