itmeJP Community


Statistics of CoS

court-of-swords

(Brugia) #21

Well and that is were an overview of all the rolls might mislead you a little bit and why I want to split it up into more detailed views. For attack rolls (one of the 2 type of rolls were rolling a 20 or 1 actually matters mechanically, the other one is death saves, which do not matter for Adam) he has a 7/8 out of 151 rolls, which is right on point and looking into even more details, by splitting the numbers into weeks we can see that Adam critted the shit out of the group in the first session of season 2, week 9, with 3 crits out of 20 rolls. Of course such variance is to be expected if you only have so few numbers.


(Brugia) #22

Currently working my way through being able to take a look at Advantage and Disadvantage. But before I finish that here the report for Week 15. Sadly I broke some of my stat aggregation, so there a few missing.

Week 15 all rolls:
Name	Average Std.    Average Success 1/20 Roll  Damage Damage Times
        Roll	Dev.    Mod.                 count Done   Taken  Attacked
Adam	 8.94	5.47	2.94	34.38%	3/0  35/1  39	  146     0
Berg	12.00	6.54	3.67	40.00%	0/1   6/0  24	   19	 10
Gideon	13.64	4.78	3.18	80.00%  0/1  11/0  46  	   20	 16
Ramus	11.82	5.49	4.27	92.31%	0/1  11/4  67	    0	  0
Snidge	 8.95	6.00	2.89	40.00%	2/1  19/0   9	    0     0

Well as the most obvious one Ramus shows off that he can actually do his own work instead of letting Adam do it.
Snidges stats are buffed up a lot by the out of combat rolls. With attack rolls being at an amazing low of 6.29 and 14% successrate, well that's what one gets for rolling everything with disadvantage. Overall though he just had an day as bad as Adam.

On JPs thought that Ramus rolled an 8 on Sacred Flame for the first time during the show, he was corrected by Dan, that Dan rolled it in the first week and Dan was actually right.
Two times was a Mogwai set ablaze by the full power of the snu cleric and in later episodes Guards avoided two times the full power of the sun. It only took Dan to repeat his feat 17 successful burnings or roughly a 10% chance for it to take this long.

Sacred Flame damage overall view:
Roll 	1  2  3  4  5  6  7  8  Total Avg.
Count	5  4  6  7  7  2  4  6  41    4.44
Success 4  0  4  4  4  1  2  3  22    4.36

(Brugia) #23

So just worked my way through Advantage Disadvantage rolls. Primarily I did them first, because I wanted to see how much they actual affect the rolls, their stats and the results the players gets if they have it.

First out of 679 d20 rolls 73 were affected by either Advantage or Disadvantage, if Adam had not forgotten, there would have been even a few more. Out of those only 23(32%) actually affected the end result. (success or fail) For the rest the player could have just rolled a single d20 and take it and have gotten the same result.
Overall there are no really unexpected things I found in the numbers, even though they are in some cases a little bit extreme, but then the count of the rolls is also really low.

	Advantage		      Disadvantage			
Name  Roll  avg mod. 	avg    changed 	Roll  avg mod.  avg    changed
      count  gained     roll   results  count  gained   roll   results
Berg	5     4.40	12.00	  1	  2    -11.00	 2.50	  1
Adam	11    7.09	13.91     2	  7     -6.86	 6.43	  3
Gideon	15    7.33	13.13     5	  2	-7.00	10.00     1
Janus	14    7.36	12.93     3				
Ramus	3     4.00	 8.00     1	  1    -14.00	 4.00	  1
Snidge	3     7.00	11.00     2	  4	-6.50	 5.75	  2
Wester	1     7.00	11.00     0	  3	-2.67	10.00	  0

avg mod. gained - Average of the difference between the 2 rolls
avg roll - Average of the d20 roll that counts for the result
changed results - how many of Roll count actually changed the end result from a fail to a success or vice versa

The most unusual thing I think is that Adam is the only one that actually got up to the expected average roll, but yet he is the least successful with his Advantage rolls.
But for the other reason why I looked at this was to see if it skews the number in a significant way. The two most obvious candidates for that would be Janus and Gideon, due to their large number of affected rolls:

Week 9-15 d20 rolls
Name	Without	With
	Adv/Dis Adv/Dis
Adam	10.20 	10.25
Berg	11.52 	11.29
Gideon	10.60 	10.98
Janus	10.75 	11.10
Ramus	10.23 	10.00
Snidge	10.4   	 8.95
Wester	11 	10.91

And yes by eliminating Advantage and Disadvantage from the rolls, those 2 are closer to the expected average of 10.5. Adam is only slightly affected by it due to his massive amount of rolls.
The rest of the group is interesting though, they actually gain in average instead of loosing it.
At least for Snidges it is easily explained. Low number of total rolls, more of them with Disadvantage then with Advantage. But Berg and Ramus have a slightly different story, not only rolled they bad on their few Disadvantage rolls, their Advantage rolls were also quite not really good, they just had not enough rolls so that it can eventually average out.
So the effects of Advantage/Disadvantage on the stats are noticeable in making them better and worse, which is interesting and one should keep an eye on it to keep the stats significant.


(Brugia) #24

So I took my time to have a look at the 4 different kinds of d20 rolls. Attack, Saves, Initiative and Skills. To see what the difference for them are. Here the interesting things I wanted to point out:

  • Attacks use generally the highest modifier, which makes sense, since they use the highest ability modifier + get proficiency bonus almost every time. Only Gideon attacked once without proficiency by throwing a rock at a zombie.
  • For most characters saves are the roll with the highest success rate, exception here is Janus, for him it was skill rolls

  • Highest crit rate for attacks is Gideon, who is tied with Wester, with about 10,5%. Directly followed with a 10% by Ramus. The lowest ranks are filled with dead and new people, Zawaz and Snidge never critted, Janus only had 3%.

  • While Berg has the highest average rolls overall, his good rolls are concentrated on Initiative and Saves, leaving his attacks and skill rolls only a sliver above the expected 10.5.

  • Most of the saves (69/86) from Adam were caused by the spellcasters of the group, with Ramus being the main cause with 60. Edit: I somehow got to 35% chance for a dc of 13 instead of the 40% it actually is, so I leave the following paragraph in, but keep in mind the difference is only 3% (which should be covered by simple variance, instead of 8%) , therefore my point is more or less null for Ramus' case
    Although Adams success rate for those saves is unusually high, with an dc of 13 he should be at an success rate of 38,1% (taking the modifier of his rolls in account). He is still a whopping 8% above where he should be, even though is d20 rolls are perfectly average at 10.51. Not sure where this comes from and I have no easy way to look at it, my best guess would be that the rolls that have positive modifiers are more often close to 13, causing the roll to be a success due to the modifier, while the rolls without modifiers take up the more extreme values.

It is even worse for Gideon, for his spells Adam saves have a 77% success rate, of the supposed 45%. Partially caused for sure by the average of 13.44 on the d20 roll.

  • But not only Adam has this weirdness in his saves. The players are even more extreme, even though I cant calculate it in detail, since I'm missing target numbers, but something way closer to 50% (probably below it) is expected, since most saves have a target number not too far above 10. Ramus and Berg are above 85%, Wester on the other hand had only 2 saves and failing those sealed his fate.

  • Skills: Janus massive amount of Skill rolls come mainly from 17 Stealth rolls, which as a rogue he should have. Taking them not into account makes him as good as equal in the amount of rolls. He also has the highest modifiers used, where he doubles Ramus, and success rate of them all. Of the people still in the group Gideon is not only the one that makes the most rolls, he also has the most success with it. Although if Snidge keeps up the trend, he might take Gideons place.

  • Initiative: Actually not to talk a lot about it, other than that Ramus and Adam are rolling kinda bad, while Berg is rolling good. Maybe a more detailed look at it is worth it, looking at Initiative position for each battle and/or how many enemies are below the character in one battle. Since alpha striking is still a real possibility.

And at last the numbers for the interested:

Attacks
Name	Average	Std.	Avg. 	Success	1/20	Roll
        Roll    Dev.    Mod.    rate            count
Adam	10.41	5.54	3.67	46.33%	8/8	177
Berg	10.64	6.08	5.24	64.71%	1/2	 33
Gideon	10.53	5.59	4.61	63.16%	1/4	 38
Janus	10.25	5.42	4.59	65.63%	1/1	 32
Ramus	11.90	5.17	5.10	90.00%	1/1	 10
Snidge	 6.29	5.62	4.71	14.29%	1/0	  7
Wester	10.32	5.94	4.68	47.37%	0/2	 19
Zawaz	 7.00	0	4.00	0.00%	0/0	  1

Saves (Snidge and Zawaz not included due to the lack of saves)
Name	Average	Std.	Avg. 	Success	Roll
        Roll    Dev.    Mod.    rate    count
Adam    10.74	5.93	0.98	47.13%	86
Berg	14.00	2.69	1.89	88.89%	 9
Gideon	11.10	5.26	1.00	70.00%	10
Janus	 9.67	5.20	0.83	66.67%	 6
Ramus	10.71	4.75	2.86	85.71%	 7
Wester	 5.00	5.66	0.00	 0.00%	 2

Skills						
Name	Average	Std.	Avg. 	Success	Roll
        Roll    Dev.    Mod.    rate    count    
Adam	8.18	4.87	0.64	36.36%	11
Berg	10.62	6.84	1.08	50.00%	13
Gideon	11.70	5.62	2.48	67.86%	33
Janus	11.60	6.29	5.93	86.21%	40
Ramus	10.23	4.82	2.68	56.25%	22
Snidge	9.64	5.32	1.73	62.50%	11
Wester	13.50	4.70	1.20	37.50%	10
Zawaz	13.33	10.69	3.33	66.67%	3

Initiative						
Name	Average	Std.	Avg.    Roll
        Roll    Dev.    Mod.    count    
Adam	 9.31	5.65	1.23	32    
Berg	11.82	4.94	1.00	11
Gideon	10.53	6.27	1.29	17
Janus	13.13	6.22	4.00	 8
Ramus	 8.19	6.01	2.00	16
Snidge	20.00	0	3.00	 1
Wester	10.00	8.54	2.00     3
Zawaz	 5.00	0	0.00	 1

Oh and some bonus trivia you could call it I guess.
Which numbers are rolled most/least for each person, some persons might not have been included fully/at all, due to their total amount of rolls making it pointless. For example Zawas would have been just a split, because he either rolled a number once or not at all.

        Least   Most       
Adam	10(20)	      25(6)
Berg	 0(13)	       7(14,18)
Gideon	 1(16)	      10(20)
Janus	 1(5)	       7(10)
Ramus	 1(6,14,18,19) 7(8)
Snidge	 -	       3(11)
Wester	 -	       5(11)
Zawaz 	 -             -

Edit: I somehow lost 5% for Adams saves somewhere along the line of my thought.


(Darkvlagor) #25

I love those proba keep up the good work !


(Brugia) #26

So I'm back again, during the content drought I was wrestling with my spreadsheets and got nothing productive done. But I don't want to bore you with technical difficulties. Also this is a little bit rushed, since it took me longer than expected and I don't want to have a half done post sitting for another half week.

First off I don't have done the live show yet, I'm waiting for the youtube vods and it will take quite a bit more time than a normal episode. (At least 8 hours, probably more to 12 with looking at the stats, picking out interesting things etc. and writing a post about it)
So if you don't want spoilers for episode 17 (the one after the liveshow) you probably should not continue reading

Now it was quite an eventful week 17, surpassing week 13 by 25% with 60 more entries, totaling at 313, it will probably be shadowed by the liveshow, which is cheating by having 8 hours instead of 4,

Once again I play around with what numbers to present, I'm sure I will find eventually the ones I will be satisfied with.
And as an usual note to things that screw with the numbers:
- I'm missing Mecha Mogwai attack modifier, which are 15 or so attacks, but they are around the normal average anyway. I know that the modifier is +3,+4 or +5; and a +4 would put them on an average roll off 10.3, so the numbers may be skewed slightly by those attacks. I think it was revealed in the liveshow but I have no time to look through it for now.
- Keep in mind that Adams rolls also contain Azure Vortexs, which is why numbers don't add up (like damage done by Adam and damage taken by the party, since he has rolled against himself and done damage to himself).

To pick out a few notable things:

  • Adams attack rolls are dumped down by disadvantage reducing it by 1.3 down from 10.8 and he lost a crit to it. For the rest of the party Advantage/Disadvantage did next to nothing. On a side note as well Adam got 4 of 6 possible successes due to bless rolls, which were all a 3 btw.
  • Across the board saves are still well above 50% and Adam was just as bad as Ramus with them.
  • While Ramus was kinda successful with his spells, his usual damage output was just not there this week. In part to low damage rolls (flaming sphere was missing around 20%) and his biggest damage aoe just getting completely saved against by Adam, although he still killed one with that.
  • Overall Adam didn't roll particularly well to crush the party, the partys rolls just were bad, but I think for the most part they just overextended their limited resources, which let to the situation in the end. One could say the 2 largest underperformers this week got what they deserved.

Week 17 Stats:

Name	Average Std. 	Average Success  1/20  Roll   Success	    Success   Damage  Avg. Dmg.  Damage	Avg. Dmg.  Times      Attacked 	     Avg. Roll of  Avg. Mod. of	 Killing 
	Roll	Dev.	Mod.	Attacks	       count  Spells	    Saves     Done    Done	 Taken	Taken	   targeted   Successfully   Adam against  Adam against	 Blows
Adam	 9.52	6.12	4.72	49.18%	 4/1   46/27  16.67%	    66.67%    188     5.53							                         8
Berg	12.56	5.83	8.00	77.78%	 0/1	9/0     -	    83.33%     87    12.43	  36    6.00	   14 	      35.71%	     10.50	   3.64	         2
Gideon	 9.17	3.97	3.17	 0.00%	 0/0	6/1   50.00%	    70.59%     21     7.00	  57	4.75	   17	      70.59%	      8.78	   5.28	         2
Ramus	 6.50	3.02	4.50	33.33%	 0/0	6/15  40%/66.67%    66.67%     75     6.25	  17	5.67	   10	      30.00%	      6.50	   4.60	         3
Persni	 9.95	5.67	7.00	60.00%	 0/1   20/1	-	    75.00%     90     7.50	  12    4.00	    4	      75.00%	     11.67	   4.08	         5

I gave Ramus 2 success rates for his spells since I'm currently debating with myself how I should count spells, which do miss damage. Left % is counting them as no success, right % is counting them as half success. Those spells are just fucking up the successes rate if you want to compare Ramus with other characters, since they are guaranteed damage. Either you can have someone who has no successes but could potentially still deal the most damage(counting them as no success) or you can have someone who deals very little damage, but has at least a 50% success rate (counting them as 0.5 success), and both options are kinda stupid to me, so you get both.

Edit: Well seems like my stat table got a little bit long there, but I'm tired so I'm sorry but you will have to deal with it.


(Typoko) #27

It will be interesting to see how the exhaustion from Bergs super rage starts to affect the numbers. For me the exhaustion from the rage is a crazy bad thing, but i'm not sure how it will play out in the end. Tho, i'm sure the exhaustion penalty will be missed quite a lot of times.


(TwoToneTerran) #28

Exhaustion isn't a big deal at the first level of it. Disadvantage on ability checks aren't really a combat problem -- you might have to do some hand holding after combat when skill challenges come in. The second level is kinda bad but manageable. The third tier of exhaustion is horrendous but should never really come up unless Berg is just constantly frenzy raging every fight.


(Typoko) #29

I agree that the first level isn't that bad. I think the worst of level one exhaustion is for initiative checks but yeah, ability checks are usually initiated by the player and most forced checks are saves, that are not affected by the first level of exhaustion.

Maybe i just have a bit too skewed opinion about gaining exhaustion. In the games i have been playing we have been using "Forced March" or some variant of it so we can get to places or get away from something that is chasing us.* It's just the worst when the long rest gets interrupted by an ambush and the party has to switch places and the exhaustion just won't go away. :sweat:

*) Athletics check to see if the party can go on longer than normal and if one fails athletics check they gain one level of exhaustion.


(TwoToneTerran) #30

Well Adam certainly loves interrupting Long Rests so that is something to worry about.


(the_pizza_party) #31

this would be much easier to do if you had access to a virtual intelligence. having to do work with your own brain is such a depressing limitation.


(Brugia) #32

Hello,
there were/will be no updates out of 2 reasons, unless I find a week very interesting.
1. Google Sheets is being a bitch and buggy as fuck and I would need to transition to something different, that I have no experience with or just write myself a little tool.
2. I started a new job in the new year, which cuts down my available free time by a lot.
For example today I spent 2 hours recording the stats (30min for the roleplaying part and 1,5h for the fight) + 3h after that for cleaning up the stats and writing this post.
But since Adam asked so nicely and I agreed with him that this session looked interesting, I sat down and did the stats for week 25.

First point to JP's comment it being the worst rolls etc.. By starting at the beginning of the session going to this point in the week, it was an average of 10.3 (17 rolls,35% Success) on their skill and saving throws. Up until the start of the fight that lowers to a 9.2 (25 rolls, 32% Success). Which is interesting because the rolls itself were not that bad, they even had 3 20s up until this point, but I think the low successrate might influence how good/bad the actual rolls are perceived.

Now for the fight stats Adam asked for, they are incredibly lopsided:

Name Average Roll Std. Dev. Average Mod. Success of Attacks 1/20 Rollcount /Caused Success of Saves Damage Done Avg Damage Done
Adam 8.48 5.96 3.81 24.07% 4/4 54/2 22.22% 99 6.6
Azriel 12.75 4.56 4.75 100.00% 0/0 8/2 / 132 12
Berg 12.08 6.26 8.23 92.31% 1/2 13/1 100.00% 170 14.17
Jubilant 14 4.8 4.2 80.00% 0/0 5/6 100.00% 105 15
Name Damage Taken Avg. damage Taken Times targeted Attacked Successfully Avg. roll of Adam against Avg. mod. of Adam against Killing Blows Avg. Roll Skill Avg. Roll Save
Adam 1 / 9
Azriel 83 7.55 35 25.71% 8.69 3.75 1 8 /
Berg 10 5 11 9.09% 8.83 3.92 5 4 20
Jubilant 6 3 6 16.67% 6.5 4 7 4 9

Edit: Well my solution to get a better table for the data only works partially, be patient while I fix this.
Edit2: Should be fixed now.
Edit3: Corrected Successrates for Adam and Berg for their Attacks.


(Darkvlagor) #33

That's really interesting and super awesome that you took the time to do this :itmejpgg:

It's incredible to see the amount of Dmg Azriel took !
He's clearly the real Tank, like, @AdamKoebel couldn't hit him for shit ! With such High AC, minions are jokes now. Our dear DM will need more than the ... +1/+2 mod. for clubing and biting ?
@DansGaming can clearly ignore minions and go for the tough, higher CR monsters in the fight. :ok_hand:

The harder thing in the 1-4 levels when you're the DPS guy, if you don't OS/2 shot a minion : You waste precious time. And the ability to already doubling the damage output is Crazy powerful as we can see with the 170 Damage by @gassy. Bergzerker is the shit, and the Brutal :itmejpcrit: is just the God Tier of min/maxing. Having a class feature lvl 9 at lvl 1 changes things.

And at last but not least the Almighty Jubilant Finisher @itmeJP with 7 Killing Blows. These crossbow shots ... Now the most interesting thing is about the Success rate. 80%+ Success, the Bless did the coffee this fight.
Also, the first round of combat has been completly soaked by the Aid spell.
So JP can have the :itmejpmlg: support title.

@AdamKoebel those rolls tho, the -10/15% to hit chance (thnaks to higher AC) did kinda help but ... I blame Canter for sucking up all your luck damn


(Brugia) #34

Bite and Club actually have both a modifier of +4, so he needed at least a 14 (for Jubilant and Berg) and a 16 for Azriel, which is rough. Although I just realized Adam(2), Berg(1) and Azriel(1) were missing hits, must have forgotten to go through the Successcolumn, when double checking my data. That does Adam only give like 3% additonal Successrate.

Bless did some work in this fight, out of the 6 would be misses, 4 times bless turned it into a hit, leaving only Berg and Jubilant with one miss a piece.