League General Discussion Thread

[gamerman01]

I have amended the league rules, effective immediately, to be clear about fighters on allied carriers (rule 10).

Basically says you need to be declaring fighters on carriers when there are allies involved. Preferably in the “game notes” feature of Triple A.

If you don’t declare, they are assumed to be on their own power’s carriers if there is ambiguity.
Declaration every turn is not always necessary because a declaration from the previous turn and the course of events dictate that there is only one possibl combination, but it is good practice to declare after each turn, whenever it is not totally obvious.

Read the rule, too, see what you think.

This rule is way overdue, even though I haven’t had disputes about this ever come to my desk.

[gamerman01]

This also means that if you intend like fighters on like carriers, you need not say anything, because that will be the default assumption, so this can save efforts and confusion, both.

[simon33]

@gamerman01 so in the scenario where there are 2 US fighters, 4 UK fighters and 2 US CVs, 1 UK CV, if I don’t say anything, the US fighters are assumed to be on one of the US CVs, with the UK ones being on the other US CV and the UK CV?

BTW, I am pretty Triple-A does track this but you just can’t see it.

[gamerman01]

@simon33 Haven’t tried hard to test it, but I’m pretty sure they’re not tracked. After all, it doesn’t ask you.
In your scenario, if the owning player makes no declaration, then 2 UK fighters are on the UK CV, and the owning player is at the mercy of the opponent for the others. Same spirit of the law as the long standing assumptions league rule

[MrRoboto]

Just FYI, some statistics for the current 2023 League

[Martin]

@MrRoboto thank you for sharing. It all looks quite equal with the exception of OOB, where the Bid should apparently be higher.

[farmboy]

@MrRoboto I’m assuming the axis win percentage by M or E players is just telling us how often do they win if when they play axis. If it is easy to do, do you have how often they win as allies? It would be interesting to know that too.

[MrRoboto]

Yeah overall BM4 and PtV are in quite a good state.

OOB has the highest bids and is still the most unbalanced.

I have one other graph:

So when top players are involved, the bid are higher and still not enough in OOB.

One thing of note: The two statistics that reference the top players (win% and bids) count games where at least ONE opponent is M or E. That’s 75 of the 163 games.

It might be more interesting to see the statistics, when BOTH opponents are M or E. However, that’s quite a small sample size: Only 16 of the 163 games are between top players…
So I don’t know how valuable that information would be

[MrRoboto]

@farmboy said in League General Discussion Thread:

@MrRoboto I’m assuming the axis win percentage by M or E players is just telling us how often do they win if when they play axis. If it is easy to do, do you have how often they win as allies? It would be interesting to know that too.

Not quite right.

That graph shows the win% of axis, when one or both opponents are M or E.
The top player can be Allies or Axis, and the top player might have lost against the bottom player!

It might also be a game where both opponents are M or E…

I can give you the exact statistic of Axis Win% and Allies Win% of Top players in a moment

[aequitas-et-veritas]

@MrRoboto Thank you for sharing!

[MrRoboto]

May I present to you: The might of our top players!

Yes, you see that correctly: Every single time a top player chose Axis in BM4, he has won! I couldn’t believe it at first, but I double checked…

For context here are the absolute numbers:

Axis wins:

Total: 44 out of 52
OOB: 19 out of 22
BM4: 15 out of 15
PtV: 10 out of 15

Allies wins:

Total: 32 out of 40
OOB: 11 out of 15
BM4: 12 out of 15
PtV: 9 out of 10

[MrRoboto]

One thing of note:

A result is only considered when the specific player is M or E in the type that’s actually played.

So if a player is M in PtV, E in BM4 but only Tier 3 in OOB, the result is only counted in the above graph when the game type is actually BM4 or PtV

[wizmark]

@MrRoboto lol sorry but i have definitely lost as axis in BM4, vs sovietishcat…who has a super strong allied game going on. Shame im the only one :(

[farmboy]

@MrRoboto Just a point of clarification as I’m not understanding what is being captured here. Is this only when the winning player is M or E and the losing player is not M or E. I read this to read that an M or E player has not lost when picking axis in BM but do know of several games where an M or E player picked axis, and played another M or E player and has lost.

[pacifiersboard]

I bet that it is about something like an IF-function? So all games of BM with a M or E player as Axis are won by a M or E player - but not necessarily the one playing Axis?

[MrRoboto]

Now I didn’t want to go into details yet, but since the issue has come up and caused some confusion, I guess I have to indulge now.

A while back @gamerman01 asked around if someone wants to be backup for the rankings spreadsheet in case something happened to him. I volunteered and subsequently got admin rights from him.

While trying to dive deep into it, I noticed some flaws with the current spreadsheet however. The most severe one: Circular referencing.

A player’s PPG determines the players Tier ranking. This determines how many points opponents get when playing that player. These points influence the PPG of the opponents and therefore the Tier ranking of all of them. The Tier ranking of the opponents however determine how many points the original player gets. This might change the tier ranking of the player and the whole cycle starts over.

So:
Player A’s PPG -> A’s Tier ranking -> PPG of A’s opponents -> Tier ranking of A’s opponents -> Player A’s PPG…

So the question is: What is the starting Tier of everyone? @gamerman01 uses past years as an indicator. However, there are some inconsistencies with new players.
Look at the overall spreadsheet right now:

Gorshak is 2-0 with 8.0 PPG, but Tier 1
jkeller is 0-1 with 4.0 PPG, but Tier M

The starting Tier is also heavily influenced by the timing and order of reults.
Use me as an example: I started this year with 4-0 and the spreadsheet therefore put me in Tier M. This meant that all of my opponents got a lot of points against me, even when they lost. Even though I have lost a lot since then, I caused a big points inflation and am therefore still E (because all of my opponents are ranked very high too because of the points inflation). @dawgoneit is another example, he started the year with a higher ranking and since he has so many games there is a huge points inflation.

Now I am a huge Excel / Google sheets nerd and I noticed all this while I was working on a side project: Currently @gamerman01 updates and maintains everything manually himself. This is a huge workload, thanks by the way for all the hours these past years!
I created an alternative spreadsheet, that heavily uses formulae and therefore is updated completely automatically!

My spreadsheet has the same problem with circular references of course. I can create dozens of different results with the same games as input!
The one I settled for is the following: All games are already put into the system (instead of one after another) and all players starting Tier is Tier 1. THEN I activate the points given for wins and losses.

I have not yet shared this with all of you since I am still consulting with @gamerman01 and want to hear his opinion first. However, he is preoccupied with real life right now so you guys have to be a bit patient.

By the way, I have made a proposal to him on how to fix these (and more) issues with the current system in the following years, but again: I want his opinion and more importantly his blessings first.

Now, @wizmark and @farmboy have rightfully wondered and questioned the above graphs. So I want to share the rankings these graphs are based on:

Overall:

OOB:

BM4:

PtV:

[MrRoboto]

Now as I said, these rankings constantly shift, whenever I make the tiniest adjustments within the sheet, because of the circular references.

I don’t change any rules, I don’t change any inputs, but this leads to different outcomes. with THESE rankings I just posted, you can see there are zero M players in BM4 and only 3 Tier E players (who have not played each other).

These Tier E players are 14-0. 8 of those games as axis, 6 as Allies.

With this data, the current graph is the following:

[Martin]

@MrRoboto thank you for the effort - I love this automation in the way you described it. I always wondered how it could be possible to manually adjust all points of past opponents and all of their opponents etc. following a new game result.

I think it is fair to take prior year’s PPG as a starting point for players who finished 3+ games and, as you suggested, 1 for new players and players who completed <3 games in the prior year.

[Martin]

PS: I also like the ELO system from Chess, but that’s a different discussion. And there is plenty of management software available.

[gamerman01]

@Martin said in League General Discussion Thread:

@MrRoboto thank you for the effort - I love this automation in the way you described it. I always wondered how it could be possible to manually adjust all points of past opponents and all of their opponents etc. following a new game result.

Not as bad as it would seem, apparently.
MrRoboto also called it impossible… It’s not that bad… :)

Busy moving into my house, will catch up later, I can’t read everything that’s going on right now, just so you all know