[thepossum]

Hi

I appreciate the specific case I am talking about was a BBO Robot tourney with us all playing South
Just a few hands and players too. 7 hands and 3 players

We all just finished on 50% and everyone missed out on the prize

Even with only 3 players the chance of that being random seemed quite low
I was feeling lazy and ChatGPT estimated around 1.148% and gave me a complex combinatorial function

I always wondered why footballers look so disappointed after a tied match

[barmar]

thepossum, on 2026-July-28, 22:59, said:

We all just finished on 50% and everyone missed out on the prize

But you all got points. Each of you was in a different stratum, so you each came in first in your respective strats and got 0.08 points.

Quote

Even with only 3 players the chance of that being random seemed quite low
I was feeling lazy and ChatGPT estimated around 1.148% and gave me a complex combinatorial function

That presumably assumes that the results are random, but duplicate is a game of skill, not pure chance. With a large number of boards you expect the cream to rise to the top, and the best player should come in first or second. But with so few boards and players, you get extreme variance.

What's more surprising to me is that there were no flat boards. And there were only 2 boards where there were 2 shared results (in both cases you shared a top with the same player), the rest had a different result for each player.

[thepossum]

barmar, on 2026-July-29, 12:50, said:

But you all got points. Each of you was in a different stratum, so you each came in first in your respective strats and got 0.08 points.

That presumably assumes that the results are random, but duplicate is a game of skill, not pure chance. With a large number of boards you expect the cream to rise to the top, and the best player should come in first or second. But with so few boards and players, you get extreme variance.

What's more surprising to me is that there were no flat boards. And there were only 2 boards where there were 2 shared results (in both cases you shared a top with the same player), the rest had a different result for each player.

I know over time duplicate shows skill differences - sad that nobody won the prize though

I wondered if anyone had even pondered such an outsome in MP before

And yes our (me and ChatGPT) assumptions had to be that every result per hand was equally likely - and it came out with some incredible maths, well beyond my care to check - just as well it was only 3 players

But getting back to your statement about duplicate, given the chance of being equal was so small does that mnean we are all the same level LOL
The hypothesis that A>B>C etc

Like in normal human type MPs I guess you need to look at the outcome where all the NS are equal, or all the EW are equual and the rare case where everyone is equal sharing tales over coffee/whisky/<insert favorite tipple> at the end of the night - so many if onlys

But yes we all got points in our Strata