1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

2019 in Numbers - ITS data analysis

Discussion in 'Access Guide to the Human Sphere' started by Ieldin Soecr, Jan 23, 2020.

Tags:
  1. Mahtamori

    Mahtamori Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    12,018
    Likes Received:
    15,302
    Yes, absolutely. I find Hunting Party to be a very interesting mission that I hope will remain in some shape or form, but what I mean is that Hunting Party doesn't require you to take specific troopers because each and every one of your troopers can complete the mission, but you can still optimize for it. Most button pusher missions don't do that, they instead require you to somewhat saturate your list with specialists with Highly Classified being the worst offender by far and I think the value of infiltrating specialists because of this has lead to the number of missions with exclusion zones shooting through the ceiling and soon the roof in the latest ITS seasons.

    But, we can do a mission review in a different topic, as pertains to this topic, missions that promote a specific type of trooper characteristic more than others will skew the statistics on what profiles are picked. But does Hunting Party really impact the value of veteran+ units enough that this trend is a bad thing, and what does missions like Supplies, that is essentially the lifeline for the low cost-efficienct skill Paramedic, do for some/most specialist skills?
     
    Abrilete likes this.
  2. eciu

    eciu Easter worshiper

    Joined:
    Nov 23, 2017
    Messages:
    4,002
    Likes Received:
    4,661
    Remove Chasseurs and suddenly MRRF will have like a 10:1 W:L ratio ;P


    Regarding MO: no surprises at all.

    Though hilarious thing is that only units with W:L ratio of higher than 1.5 in MO are Trauma-Doc and Palmbot xD
     
    theradrussian likes this.
  3. Hecaton

    Hecaton EI Anger Translator

    Joined:
    Nov 23, 2017
    Messages:
    7,205
    Likes Received:
    6,535
    I don't think he can vault mid-jump.
     
  4. Mahtamori

    Mahtamori Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    12,018
    Likes Received:
    15,302
    I have unsuccessfully made that line of questions in the past, generally speaking it doesn't seem like people are willing to accept this fairly massive slow down of play.
     
  5. Hecaton

    Hecaton EI Anger Translator

    Joined:
    Nov 23, 2017
    Messages:
    7,205
    Likes Received:
    6,535
    I've definitely enforced it. Now, what I mean is, he can definitely jump through things he could vault over on foot, but he can't just jump up to a building until the top of his silhouette reaches the roof and then vault to the top for free.
     
  6. Mahtamori

    Mahtamori Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    12,018
    Likes Received:
    15,302
    Right. No, that's absolutely out of the question, and not really what I was suggesting. I was instead suggesting that when jumping up a building, people suggest that the railing would be vault-able. By that logic you don't need to clear the entire height of a fence if jumping a fence, so kind of like instead of using Scissor technique you'd do a Fosbury Flop. And what isn't a building to McMurrough but a somewhat thicker fence?
    So I'm suggesting that McMurrough can "flop over" a building, not "flop up" it, if it is at all possible to vault during a Jump.
     
  7. Brokenwolf

    Brokenwolf Well-Known Member

    Joined:
    Mar 8, 2019
    Messages:
    1,184
    Likes Received:
    1,832
    I do not know if this is possible, but did anyone look at this data and analyze what was the average order count per faction?
     
    Xeurian and emperorsaistone like this.
  8. ijw

    ijw Ian Wood aka the Wargaming Trader. Rules & Wiki
    Infinity Rules Staff Warcor

    Joined:
    Apr 25, 2017
    Messages:
    7,331
    Likes Received:
    14,817
    I don't think that can be extrapolated from the spreadsheets that @Ieldin Soecr uploaded, but it'd definitely be interesting. Both in terms of Trooper numbers and Regular Order numbers.

    Also, whether there's any correlation between force size and win rate.
     
    Xeurian and Brokenwolf like this.
  9. Ieldin Soecr

    Ieldin Soecr Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    129
    Likes Received:
    329
    Average unit count per Faction and Sectorial, including order count divided by regular/irregular/impetious, as well as the win rate.
    Data are for reported ITS10 and ITS11 events until May 2020. The unit count includes non order generating units (Servents, etc.). Aleph and OSS order count is around 2 orders to high, as proxies count as 1 order each, not one for all 3

    Data Download
     
  10. Brokenwolf

    Brokenwolf Well-Known Member

    Joined:
    Mar 8, 2019
    Messages:
    1,184
    Likes Received:
    1,832
    So on average, it looks like Dahshat, Caldonia, and Ikari are the groups that really not liking the change (average of 17 and 16 orders), but that number includes Servants.

    I wonder if CB made the decision based on similar data?
     
    Jumara likes this.
  11. Hecaton

    Hecaton EI Anger Translator

    Joined:
    Nov 23, 2017
    Messages:
    7,205
    Likes Received:
    6,535
    I think that a legitimately big part of the decision had to do with round timing more than balance.
     
    Jumara and meikyoushisui like this.
  12. Teslarod

    Teslarod when in doubt, Yeet

    Joined:
    Nov 23, 2017
    Messages:
    2,407
    Likes Received:
    4,864
    Hm as usual those numbers look a lot more realistic than whatever CB shows at events.

    Since we all know what you guys want to know:
    by WR
    1. Dashat - 63.52%
    2. Aleph Vanilla - 62.14%
    3. Spiral - 61.37%

    3rd from last. QK - 41.61%
    2nd last. PanO Vanilla - 40.73%
    dead last. Druze - 39.63%

    Full list by WR
    upload_2020-8-18_23-6-37.png

    Average Units per List
    14,15
    Average Regular Orders per List
    12,03
    Average Irregular Orders per List
    1,56
    Average Impetuous Orders per List
    1,39

    Surprises for me:
    Nomads "worst" overall Faction (purely by WR)
    PanO Vanilla 3rd last
    Spiral Top 3 (would have guessed 55% ~top10)
     
    #92 Teslarod, Aug 18, 2020
    Last edited: Aug 19, 2020
    Ariwch, Lesh', meikyoushisui and 2 others like this.
  13. Triumph

    Triumph Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    5,715
    Likes Received:
    6,472
    One request, could you present important factional data such with an extra graphic where only larger events are counted? As much the smaller events are very important to the lifeblood of the game, they don't provide great data for analysing competitive results.

    One thing that might be fucking with your results here is that the special ITS seasonal packs lock the missions but they are very attractive to TO organisers because of all the prizes and minis inside (like the new TAG sculpts). Personally, when the seasonal pack drops I buy a whole stack of them (like 5 or 6) and just use codes from those. Our actual events use different missions but I can't change them in OTM to reflect this.

    I mentioned this to @HellLois and @Koni and suggested they unlock the missions because the end result for them was TOs were just going to fuck with their data, but nothing was done about it.

    EDIT: In fact I have a season X event set for October, missions are locked at Frontline, Aquisition, and Supplies but those aren't necessarily the missions we're going to play.
     
    #93 Triumph, Aug 18, 2020
    Last edited: Aug 18, 2020
  14. Armihaul

    Armihaul Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    1,470
    Likes Received:
    1,112
    Thanks!. Apart from winrate, we need also average TP aqquired. It is not the same winning by 2 or by 3, and that helps to valorate ties too. Also, can you put some context? is this all from 2019? only previous ITS?

    Some of us have been telling something like this for some time, is just that people don't want to accept it. Nomads were once one of the big fish, but not anymore. Seems that in L.I TJC still kicks in (if we look at CB's data from some months ago), but even I didn't expect so bad results. That's why I asked for some context, because it contracits some of what we saw last time.

    taking into account only the "big events" creates a bias. Because competitive players tend to alter their list to that tournament (some might chose different factions for a tournament), but also, because there will be taking out both of the "not so good players with the op faction" and "the good player with the bad faction". The bigger the imput data, the better results and more acurate. Also, we cannot confirm that the cuality of the players is more equivalent in a big tournament than in a small one. I have played even in interplanetario (and I am not a good player) so why my results should be taken into account, but results from other people that might be better than me, but didn't have the chance or time to go to those events shouldn't?
     
  15. theradrussian

    theradrussian Well-Known Member

    Joined:
    Jan 4, 2018
    Messages:
    506
    Likes Received:
    851
    If this isn't a damning indictment of the state of MO, then nothing is.

    Oh wait, now for me to be hit with some WeLl AkShUaLlY
     
  16. Armihaul

    Armihaul Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    1,470
    Likes Received:
    1,112
    thinking about this expreselly, it could be insteresting to split the results between players results. I mean:

    picking the 1/3 of best players, do the previous maths.
    picking the 1/3 of worst players, do the maths again
    pick the remaining players, do the maths and then compare

    it could also be achieved looking at the top half, the botom half, and doing a subtraction (top minus botom) between the %, but that brings less info.

    with the complete data it could be possible to achieve. That might be an interesting look to see "the balance" between factions in the way "those are easier to get better results than those others", or "those are chosen more by top players"
     
    bladerunner_35 likes this.
  17. Teslarod

    Teslarod when in doubt, Yeet

    Joined:
    Nov 23, 2017
    Messages:
    2,407
    Likes Received:
    4,864
    Here's the tierlist by TP for 2019 again:
    1. 0.54 Dashat
    2. 0.52 Shasvastii
    3. 0.51 Tohaa
    4. 0.51 Spiral Corps
    5. 0.51 OSS
    6. 0.51 Aleph Vanilla
    7. 0.51 Varuna
    8. 0.48 Combined Army Vanilla

    9. 0.48 Caledonia
    10. 0.48 SAA
    11. 0.46 Ikari
    12. 0.45 Foreign Company
    13. 0.45 TAK
    14. 0.45 FRRM
    15. 0.44 Steel Phalanx

    16. 0.44 Ramah Taskforce
    17. 0.44 Hassassin Bahram
    18. 0.44 IA
    19. 0.43 Nomads Vanilla
    20. 0.43 Ariadna Vanilla
    21. 0.43 Neoterra
    22. 0.42 Starco
    23. 0.42 Bakunin
    24. 0.41 MO
    25. 0.4 JSA
    26. 0.4 Onyx
    27. 0.4 Tunguska

    28. 0.4 Haqq Vanilla
    29. 0.39 ISS
    30 0.39 USARF
    31. 0.38 Corregidor
    32. 0.37 Morats
    33. 0.36 PanO Vanilla
    34. 0.35 QK
    35 0.35 Druze


    All in all very similar to this data sample's spread in WR, seems like the metrics are fairly comparable. Winning is by far the biggest swing as ties are relatively rare and win<->lose is a 2TP difference.
    Context was mentioned. The new data set is all of ITS10 and 11 so far til May 2020.

    Remember this is overall data. So AVERAGE relative performance.
    I'd heavily disagree PanO, a Vanilla Faction with 5 Sectorials to draw troops from is bottom 3 in potential. It's easy over interpret the data with what isn't there.
    For Vanilla Factions with loads of choices bad performance mostly means factors that pull down the average. For PanO that has always been the composition of Icestorm leading to a lot of people trying their first ITS game with the box they have. It's also relatively hard to build a competent PanO Vanilla list, there are loads of duds and traps on the way to include all the tools you need. Same for YJ, Haqq and Nomads, you notice none of them perform particularly well on average.
    Aleph and Tohaa on the other hand have a lot of very viable and competitive choices, it's much easier to end up with a competent list that covers all bases without even trying there.
    For Sectorials with a fraction of the choices, this tierlist is much more accurate. There's no 5 TAGs and 5 Elite HI to pick from and tryout. If a staple unit isn't quite it's less likely to be a matter of composition but instead hints at internal balance problems.

    If we had perfect data we wouldn't have to have a discussion. With 10 million games played per faction each season we could just check the performance for the top 0.1% and netlist a meta. That still wouldn't work as easy as that, you still need to consider missions and matchup and a 2 list format, but at least the baseline would benefit.
     
    meikyoushisui likes this.
  18. Armihaul

    Armihaul Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    1,470
    Likes Received:
    1,112
    thank you!

    so I assume then that Limited insertion, tactical window and no-extra games are mixed together?

    the comparison between TP and winrate is to put value for the winrate and ties. A faction with a high winrate but low TP average could mean that they win by little and tend not to tie, while a lower winrate but with higher TP could mean that is a hard to win with army, but also hard to win against
     
  19. Tristan228

    Tristan228 Bakunin's best Morlock trainer
    Warcor

    Joined:
    Nov 23, 2017
    Messages:
    1,168
    Likes Received:
    2,217
    Nice work. But wouldn't it be more informative if you added the standard deviation to the average values and the total games played per faction? Also don't Limited Insertion and Tactical Window (since both were/are extras in those seasons) skew the data?
     
  20. HellLois

    HellLois What the Hell...Lois?
    CB Staff

    Joined:
    Apr 25, 2017
    Messages:
    1,537
    Likes Received:
    4,137
    Nice, very intereting, conclusions you got, Like us :D So, it sounds like a good idea to limit to 15 troops, oh wait :P

    [dont worry, we also said thet there will be a FREE MODE on ITS, without that limit ;) ]

    Yep we took note of that. With the next event pack code you will be able to create the tournament as you want.
     
  • About Us

    We are a company founded in 2001 in Cangas (Spain), and devoted to design and manufacture games and figures. Our main product, Infinity the Game, was born with the ambition to satisfy the most demanding audience, offering the best quality.

     

    Why are we here?

     

    Because we are, first and foremost, players.

  • Quick Navigation

    Open the Quick Navigation