Leagues Stats - Season 7/8
Most popular team event vehicle parts
So I put together the chart below. Basically I took all Team Event Strategy Info from SoulisBot (thanks Soulis 👍) from about the last 6 months or so and counted how often each vehicle part was mentioned. I sorted the vehicles from most popular (Hotrod) to least popular (mono) as well as for each vehicle, which parts were used most - which I listed from left to right (e.g. Coin Boost was the most popular for Hotrod). I left out any part that was used less than 15% of the time to shorten the list. Overall, Wings was clearly the most popular part followed by a close competition between CoinBoost and Afterburners.
Average Team Scores per round at the boundaries for Promotion and Relegation (Season 7)
(assuming all matches played - Thanks to Apru for taking snapshots of the final rankings)
- The promotion boundary from Division 1 to CC dropped by another 40 points as teams increasingly attempt to avoid promotion
- It is still easier to enter Division 2 by being double-promoted from Division 4 (2529 average score) than to be single-promoted from Division 3 (2770 average score) but has narrowed a bit. Similar story for entering Division 3 or 4.
- Estimated total scores for promotion and relegation in December (Season 8) are shown in the second table
Did teams in CC & D1 around the #50 rank sandbag to avoid CC for Season 8?
The quick answer is not really for CC, but very much so for D1!
I compared the increase in each rank’s score with the prior average. Assuming teams played as normal, the two should not be that different, although the average should be more stable. And that is exactly what we find in CC.
However, in D1, the last match scores are substantially below that of the prior averages for teams ranked between 25 and 100. The top 25 seemed content to be promoted or perhaps resigned to their fate as they played as normal. However, below this, many teams either skipped their last match or did not put in a full effort.
This was especially clear between ranks 35 and 60 where the increases were 2000 or less despite those positions having previously average about 3000 per match. The nadir was #41 which only increased 1000 suggesting that it was teams ranked around #40 or so that realized that they had to take extreme action to avoid promotion.
Pedal+Metal Virtual League. Enter your scores to see how you would compare against NORTH. https://docs.google.com/spreadsheets/d/1TrdG6Pfpjz-8PKp9zRcspK4JfvnCur8amJoKM7-zfCM/edit
A lot of strength at the top of ?. 2nd quartile of Germany appears stronger. And the rest is very even. (The ones at the end don’t really count)
Thor's Hammer-2 22240
Thor's Hammer-1 18605
Gas Gas Gas Virtual League https://docs.google.com/spreadsheets/d/1qZY05571CFfh-V6oy4RXw-3cIYozHryl4Y75dkV1ZDg/edit
I entered the ? ? And GERMANY scores into the Gas Gas Gas virtual league sheet. I don’t know what is more impressive? That both teams averaged 35k or that both had full teams with 100% participation! Interestingly, Germany had a higher average score but still lost by 400 points, which really goes to show how important those top few places are (? ? Had the top 5 spots) https://docs.google.com/spreadsheets/d/1qZY05571CFfh-V6oy4RXw-3cIYozHryl4Y75dkV1ZDg/edit Feel free to enter your team’s scores (on the input sheet) to see how you would have done against them.
GERMANY scores much more consistent. “? ?” Has some really strong players (alt’s from their first team ? ) but does not quite have the depth of players that GERMANY has. Team scores in the chart below ↓
How much to teams improve between matches of the same event On a separate note, based on a study of Div 1 teams, players increased their scores by an average of 1k for every new match of Squad Goals.
Change in #Teams since last season
Season 7 - Final Results
- Div 1
Top 305 in November
- Div 2
November Top 1-205, 1197-1205, 1598-1606
- Div 3
– Source: Apru
- Div 4
– Source: Apru
- Div 5
– Source: Apru
- Div 6
– Source: Apru
- Div 7
– Source: Apru
Leagues Stats - Season 6/7
Comparison with Division I and II
- Although no superteams in I and II, clearly relegated teams from CC and I respectively are dominating in their new (lower) division with scores in the 3000-4000 range.
- Division I remains the best balanced with most matches in the 2000-2500 range, but not by as much as before
- Doubly promoted teams from Div IV still resulting in a group of relatively weak teams in D2 with scores < 1500
D1 first match
- It really stands out how many close matches (2000-2500) there were in D1.
- My team’s match included. 2261 vs 2259 (we won 😅)
D2 first match
- A fairly flat distribution suggesting quite a bit of divergence in team strength
CC Season 7 - First Match
- A fairly uniform distribution with thick tails. Probably reflects the randomness of the matching as well as the enormous divergence in team strength resulting in the thick tails.
- That there was more teams that scored >4000 than 3500-4000 indicates the shear strength of the super-teams!
CC Season 6 Final Results Distribution
- Clearly 5-10 teams that really didn’t even try
- interesting that there were fewer teams in 2000-2500 range than 1500-2000 and 2500-3000. I think this represents not only the divergence between the Top 50 and Bottom 50, but also that teams were avoiding the promotion boundary by taking steps to ensure relegation (not playing, say) or avoid it by lining up an easy last match.
Here is the latest Pumpkin Juice Virtual League Spreadsheet. Feel free to enter your team’s scores in a free column to see how you compare… https://docs.google.com/spreadsheets/d/1nI1-_3-H1uf-L7vzPZAcHgSVyw4MUiwPV2u4ezc3ZyM/edit
Average Team Scores per round at the boundaries for Promotion and Relegation (Season 6) (assuming all matches played - Thanks to Apru for taking snapshots of the final rankings)
- Note due to a bug, it was not possible to record boundaries for Divisioms 5, 6 & 7. Accordingly numbers for these divisions are based on Season 5 with an adjustment for the extra match
- The promotion boundary from Division 1 to CC dropped by 40 points as teams increasingly attempt to avoid promotion
- It is still easier to enter Division 2 by being double-promoted from Division 4 (2378 average score) than to be single-promoted from Division 3 (2707 average score)
- Estimated total scores for promotion and relegation in November are shown in the second table
Here is a link to the spreadsheet if you are interested. You should be able to input your player scores in columns D:AA (one column per team) … https://docs.google.com/spreadsheets/d/1uFH1-HRvjmFWrnwRnj4jg_iJnb_Zbn1AY__btsOw8sk/edit?usp=sharing
My team tried to line up a match with Thor’s Hammer - the only US flagged team ranked higher than us in our division, but it didn’t work. Nevertheless, we are now running a virtual match where we each input our player scores into a spreadsheet and it tells us the virtual team score! It is working pretty well (other than my team is behind). Current score:
Gas Brake Go 2132
Thor's Hammer 2367
The spreadsheet can also handle more teams via two methods:
1) Virtual League : Total score from all pairwise virtual matches
2) Super Match : Single match with multiple teams participating
Thought I would mention it here in case any teams out there are curious how they would do if they matched another team. Note: it does require inputting the individual player scores.
Season Change in #Teams
Season 6 - Final Results - C.C.
– Source: Apru
Season 6 - Final Results - Other Divisions
– Source: Apru
Season 4 - Final Results - Div 3+?
– Source: Apru
Leagues Stats - Season 4
How badly did teams want to be promoted and avoid relegation in Div. 1 last season?
The quick answer is not at all when it came to being promoted! Division 1 teams know that facing super-strong teams in CC is brutal, and may be willing to take extreme steps to avoid that fate. One way to see this behavior is to analyze the last (15th) match of the season - or lack therof! But I’m not going to go into too much detail - just give you a taste of what happened last season.
Let’s start with the relegations from Division 1 (D1). I think this performs a decent control group as D1 teams surely do not fear D2 teams; nevertheless, they are likely to wish to remain in D1 for the better rewards. Any team ranked 301-500 at season-end is relegated. After 14 matches the relegation border was 29036 points (an average of 2077 per match). After 15 matches, it increased to 31149 points. 31149-29036=2113. So a touch higher than the average indicating borderline teams did make an effort to stay in D1 (using extra tickets perhaps?). And probably no borderline teams in the region tried to perform poorly or skipped the last match in order to be relegated.
However the story was quite different for promotions to CC (top 50 teams in D1). After 14 matches, the border was 42414 (an average of 3030 per match). However the final promotion yardstick turned out to be 43755 - only 43755-42414=1341 higher. So teams in that region that were averaging over 3000 per match averaged well less than half that for their last match? The inference is that more than half the teams at risk of promotion skipped their last match to try avoid being promoted!!
Season 4 - after 10 Matches
It’s amazing how similar the distributions of average scores for the top 3 divisions have become. Division 1 remains the best balanced with most fair matches and very few walkovers. Both CC and Division II do suggest a greater number of weaker teams (more teams in the left tail) - but these are looking better than previous seasons. In fact this is true of Division 1 too, where the tails are smaller than previous seasons (fewer super-strong and super-weak teams). Why is this? Well there are two reasons that come to mind…
1. Teams are gradually being sorted into their right divisions. Prior to leagues there were many strong teams that were resets to farm better rewards. With season bonus awards providing an incentive to being in a higher division, many of these teams have been working their way up to their correct divisions. There were also teams that were highly ranked before due to a series of lucky matches with weak teams. These teams have not been able to compete in the high ranks and have fallen back to lower divisions. Both of these should result in fewer extreme match results.
2. Teams are doing it for themselves. Many who have tired of unfair matchups and do not wish to risk relegation (or promotions from Div 1, say) have formed alliances with teams of similar strength to ensure more competitive matches. This is especially prevalent in CC by the looks of it, given the dramatic change in average score distribution to being much more like the other divisions! It is quite ironic that the lack of fair matches that leagues failed to provide now seems to be indirectly succeeding via concerted team action to address it.
CC Only Discord and Not Redd|It have managed to average over 3500 per match, but 15 teams have averaged over 3000. The boundary average score required to be in the Top 50 (out of the relegation zone) is almost 2300, which is very high. Perhaps teams at the bottom not really trying and gifting lots of points to their opponents?
D1 Only the Portuguese speaking team, BR & PT, averaging over 4000 with another 10 teams averaging over 3500. Top 50 promotion boundary just over 3000, but may come down assuming many teams don’t want to be shark-fodder. Relegation boundary for the bottom 200 is almost 2100 average score!
D2 Top 200 promotion boundary at almost 3100. Double relegation boundary at 1500 and single relegation at 2050. 3 teams averaging over 4000: PRO-TAGONISTS, Not Redd|It² and OLDSCHOOL 2.0 - and another 21 averaging over 3.5k.
Season 4 after two matches
CC: Bimodal distribution suggesting very different strength level between Top 50 and Bottom 50
D1: Remains a well-balanced division with a very normal and symmetric distribution of scores
D2: In-between with well-balanced teams but many weak teams too
Season 4 - Final Results - C.C.
– Source: Apru
Season 4 - Final Results - Div 1
- source NKRI
Season 3 - Final Results - C.C.
- I T A L Y² 826 – Source: Apru
Season 3 - Final Results - Div. 1
– Source: Apru
Season 3 - Final Results - Other Divisions
Leagues Stats - Season 3
How well are teams that were promoted doIng in C.C.? Promoted teams are widely regarded as Shark Fodder to be chewed up and spat back out to Division 1 after one season. Is this true?
Well I took a quick look at the Top 50 teams in C.C. to see how many had made there way out of the relegation zone. I found that there were ten from Division 1, placed as follows: 13, 24, 29, 30, 34, 41, 45, 46, 48, 50. Half of these were still outside the Top 40.
Accordingly it appears that of the 50 promoted teams, 40 are likely to be relegated after one season, 5 may just hang in for another season or two, and 5 may be longer term C.C. residents!
After Match 10
Season 3 - C.C. - After 10 Matches
Very peaked distribution of average scores, with thick tails (I call this the Middle Fingersoft Distribution 🤣). I.e. lots of real strong/weak team and some in the middle too - with few in between.
Projected Relegation Score for 15 matches is 33900 (average of 2260 per match)
Season 3 - Div 1 - After 10 Matches
A very typical bell curve. Only one team averaging over 4000 per match (Norden Elite). Small tails suggesting fewer walkover matches.
Assuming 15 matches, projected score for promotion is 46500 (3070 per match average), and a total score of 31125 or less to be relegated (average of 2075)
Season 3 - Div 2 - After 10 Matches
A bit of a scewed distribution suggesting many weak teams and few super-strong teams. Four teams with averages above 4000 per match headed by TÜRKIYE. Assuming 15 matches, expected total score required for promotion is 46200 (average of 3080 per match). Double relegation est.score < 23500 (1570 average) and single relegation est. 30k score (2000 average)
Comparison of Division Distributions
Extent of mismatches clear in C.C. given thick tails
Matches are closest in Div 1 with more consistent average scores in general
Div 2 super strong teams as scarce as in Div 1 - but a heavier preponderance of weaker teams (promotions from Div 4?)
After Match 1
- After the first match in C.C., the median score was 2259. This will probably reduce over time assuming some teams are not able to match in future rounds.
- It is noticeable how uniform the results were spread. A function of this being based on a single, randomly-allocated match.
- Italy² came last with 30 cup points - which is not surprising as they are down to 3 members and the leader has <2k gp. Presumably the members are off to better hunting grounds for the season. How many other teams will do the same? They are unlikely to be the only ones. It looks like GERMANY was the lucky team that benefited from the easy match!
- With one or two exceptions, the vast majority of lower ranked teams (e.g. 8 of the bottom 10) are imports from Division 1. Welcome to C.C.!!
- The distribution of scores in Division 1 is quite Bell Shaped with most teams scoring between 1000 and 3500. A sign of very little variation in team strength.
- 499 teams in total. The bottom two are clearly inactive with 3 player between the two of them. The unlucky team that was unable to find a match was GermanEmpire.
- 1997 teams of which 23 did not match
- The triangle-shaped distribution suggest less variation is strength than C.C. - but more than Div. 1.
About 30% of Division 1 matched within a 5-10 minute window. This indicates that many many teams joined the season right as it started. I wouldn’t be surprised if this has something to do with teams missing out on the 15th match in Season 2?
With so many teams starting around the same time, I’m hoping that the matching algorithm will do a better job🤞
Season 2 - Final Results - C.C.
100th team ЖУКОРАМА with 0 trophies.
– Source: Apru
Season 2 - Final Results - Div. 1
– Source: Apru
Leagues Stats - Season 2
Final Analysis for Season 2 - After 14 Matches
The analysis is based on results after 14 matches as many teams missed the matching deadline for the 15th match, and I did not want the results distorted by this.
A few observations:
- No teams averaging more than 4000. This suggests that even the weakest teams have 1 or 2 decent players.
- As many teams 3500- 4000 as 3000-3500. This lack of tailing-off indicates the population of 10+ super-strong teams that almost always do well.
- There is a big jump in the number of teams from those scoring 0-1000 to those scoring 1000-2000. I would imagine there are about 10 teams that have not even been trying - with players departing to other teams only to return at the end of season?
- The dividing line between top 50 and bottom 50 is approximately 2180 average score
- Fairly large proportion of teams scoring between 1500 and 3000 suggesting a lot of competitive matches
- A few very strong teams (3) averaging above 4000 that should do fine in CC. but the rest may struggle more.
- Probably 100 teams that belong in Division 2 but another 100 being demoted that will dominate in Div 2 and be back the next month.
- Teams averaging less than 2000 will be demoted.
- Teams averaging above approx 3000 will be promoted
- Very few teams scoring above 4000 suggesting a lack of C.C. level strong teams.
- Only 50 or so teams averaging above 3500. The other 150 being promoted may struggle in Div 1.
- Fairly normal looking histogram otherwise
- Average of about 2425 or above required to be promoted,
- Average of less than 2000 to be demoted, and less than 1440 to be double-demoted
Comparison between Division Distributions
Season 2 - 11 Matches
- After 11 matches, the relegation demarcation is roughly 2215 average score
- For C.C., there are more teams averaging 3000-3500 than 2500-3000, which may indicate two separate population groups. I.e. the super-teams vs the rest. This property is called heteroscedasticity. And that is your stats lesson for the day 😁
- So after 11 matches, it looks like relegation for those with <2040 average score, and promotion for those averaging 3150 or more (roughly)
- it will be interesting to see if this changes going into season end - or if some marginal teams will decline to play.
Season 1 - Final Results - C.C.
- Source: Nemesis News
Leagues Stats - Season 1
- 10 abandoned teams!
- Max points is 4526 for 100 players, split between the two teams. However. many teams are not full (and a few may not match) so total points will be less on average, say 4500. Accordingly averaging at least half this (2250) should be enough to not get demoted (but no guarantee). Therefore, assuming 15 matches over the season, a total of 2250x15=33750 may be enough to be in the top 50, and thus avoid relegation. Best of luck!