Wide Payout system (for round-robin tournaments)

[Peter Bäckgren]

I decided this is the most appropriate site for publishing the Wide Payout system we used for years in the Finnish League. Jacob has some interesting ideas and I was discussing some of them with him over a year ago BUT then I ended up building a house and getting a kid at the same time and life just basically completely shut down anything backgammon related :(. But better late than never.

Firstly, I was impressed by the fresh ideas in Jacob's book (Fair Backgammon if he has written more since). Unfortunately I don't have the book handy so I'll just refer to my notes before diving into how the Wide Payout works. Jacob had some fixes for tournaments that had a "better" way of calculating winners. If I remember correctly you got points based on some clever but a bit complicated formula. What I wrote down when reading the book was that it would at least theorethically be possible to manipulate the results using that system. i.e., a player not faring so well could start "surrendering" towards the end of tournaments which would skew the results.

Well, I had the same problem (both theorethically and in practice, albeit not by players on purpose) in the Finnish League I was running for several years.

The tournaments were run using a round-robin system which has the obvious flaw of players sometimes dropping out once they have no real chance of winning. It makes little sense to ban players if they do not finish all their matches properly and that would also invite people to basically do forfeits.

Well, if the stick does not work, always try a carrot instead.

In backgammon, the carrot is obviously money :D.

So, instead of using a common 70/30 or 60/40 money split I devised a payout system that gave money back to most of the players, a Wide Payout system. In practice this meant the winner got 50%, the runner up 25% and the remaining players split the 25% on a declining basis. This is best explained with a practical example of 10 players where the price structure would be (roughly) the following:


1 :44.5%
2 :24.5%
3 :8%
4 :6.5%
5 :5%
6 :3.5%
7 :2%
8 :1%
9 :0%
10:0%

Never mind the percentages being so odd, this was interpolated from a 20 stake, 10 player actual payout (i.e. 200 => 100%).

It is probably possible to devise a clever formula for this but then you get into trouble with tournaments with very few (or many) players. At the end of the post I have tables for 5-11 players. The two last players are always left without returns. However, the purpose of this post is not to give exact details, but to tell you the benefits of using a Wide Payout.

One of the reasons I started the Finnish League was to give average players a chance to complete for affordable entry fees. A Wide Payout is not all that attractive to top class players as the payout is less than what they are used to (at least if they win). In the end there was a fair amount of top class players playing in the Finnish League so apparently a good challenge is rewarding in itself as well.

To an average player the Wide Payout gives more reason to strive for excellency in every game. Losing a single game can mean the difference between position 2 and 8 where position 3 commonly roughly gives you your stake back. For those of us having a wallet smaller than a building brick staying even means you will probably play more games on average per year.

Also, by having a wider payout players can expect to get some of the original stake back. Thus, the entry fee can be higher, on average another 50% can be added. Which of course makes every position count more money-wise.

Even if top notch players might not care, they get some benefit from the system since a bad luck streak mostly only lowers their total losses instead of the normal situation of "lose the two first games and you are bust, all remaining games are irrelevant". Similarly a below average player still has fun left in his last games trying to lower his "entry fee".

So if you want to devise round-robin tournaments, this is not a bad system to use. Even with 3p matches the tournament become more hectic when every match counts (99% of the time) all the way to the end. In case you use this system, a mentioning of Eskimo as the author is appreciated but not mandatory.

It is possible I forgot something important since it's a few years since I was running this. So criticism is welcome, faults might just be an omission since this worked well for several years after all.

Also, if someone finds this interesting I can go a discussion about how to implement this in other tournaments as well.

-----------------------------
Selective dump from the original rules used. Mainly added to this post because you can see an actual example of how the winnings are paid out.


Each player plays against each other player in his division.

The winnings are awarded using a Wide Payout System. Most of the cash is divided between the two first standings but a considerably
smaller amount is divided between all other standings except the last two ones. In reality this means the entry fee is somewhat lower
for everyone on average. From experience a Wide Payout leads to every point being important which leads to all matches being actually played.

Win : 3p
Loss : 1p
Forfeit : -1p

Standings based on points, then internal victories if even and finally match scores if round-robin even ties.

---tables---

20e, two last don't get anything, otherwise close to 50%, 25% and the last 25% divided between 3-(n-2)

5 (special case):
60
30
10
0
0
--
100

6:
70
35
10
5
0
0
--
120

7:
77
39
12
8
4
0
0
--
140

8:
80
40
16
12
8
4
0
0
--
160

9:
90
45
15
12
9
6
3
0
0
--
180

10:
99
49
16
13
10
7
4
2
0
0
--
200

11:
110
54
14
12
10
8
6
4
2
0
0
--
220

[Jakob Garal]

Hi, Peter! Thank you very much for your proposal. It is very interesting that you have very closed idea to FPP-tournament I have implemented about two years ago at Kharkov Federation of Nard and Backgammon.  Now I have developed it further and we have new format of backgammon tournament with prize fund sharing. You can find more description and results in First Ukraine Open Online, that we are promoting at GRIDGAMMON. You can take part in autumn session that will be in November.

There are some comments of backgammon profis Mochy and Michy too.

Your comments to my book I'll answer you too - sorry for late answer.

According to sensor board "BEZMA" - your remarks and comments are welcome => [e-mail]

[Peter Bäckgren]

Ok, I agree that the preliminary game round winnings you have used is trying to address the same problem I was aiming at with the Wide Payout.

For a Swiss tournament your payout seems very comfortable and hopefully will gain some acceptance.

For a round-robin league like the one I devised using the Wide Payout, WP might or might not be better.

WP Pros and cons compared to Prize Sharing? (Not sure what you call it):

Needing a forfeit penalty in the WP game system to force each match to be played is not "clean". With 1 point already being assigned to a loss in the system the PS could work as well to achive a similar result. Will consider if I ever start running tournaments again.

Calcing PS wins/losses might take more time than in WP (albeit it's probably just an excel table or something). WP shows the entry vs price allocation directly (although the allocation is less clean and ugly for the arranger if the playing field is larger than existing precalced winning tables - albeit only once :).

Technically PS could give nothing (or like 1 win max) to certain players if the playing field is very uneven. But this is theorethical. Since WP has players playing each other having lots of players with "only losses" is not possible. What I'm trying to say is, on a *very theorethical* level WP might work better for round-robin if the playing field is badly stacked like 2/3 pros vs amateurs.

WP is slightly stronger in shifts, i.e. one win could push you quite a lot higher in the charts compared to losing the game. Obviously this is the effect of having win/loss rewards of 3p/1p instead of the swiss 1p. But the net effect was often that quite many players could reach 2nd position (with it's higher payout) even if the overall winner was clearly decided early on. Another benefit is you sometimes wanting to watch another game after you have played all your matches as it could severely impact your own position.

These two effects, IMHO, are clear benefits compared to a swiss tour where the shift always is minimal (in the final round(s)).

So in the WP vs PS I'd call your system the winner in Swiss formats and quite good for Round-Robins (which can only be small-scale anyway). While writing this I was contemplating switching to PS myself for round-robins but then I convinced myself otherwise when remembering the "fun" at the end of a WP tournament when the two last games of the entire league were nailbiters for almost every other player as well. And nailbiting was quite common during the 20-30-40 (?) tournaments I arranged using WP.

But short on time so I'll have to get back to the BEZMA later.