You need to be logged in to your Sky Poker account above to post discussions and comments.
You might need to refresh your page afterwards.
** JACKPOT SYNDICATE ** FLAWS IN THE REMOVALS SCORING SYSTEM ** PROPOSALS FOR 2018 **
StayOrGo
Member Posts: 12,181
Hi all.
There are a few flaws in the scoring system for the removers system and being a poker player, I am always looking to sort out leaks. :=)
The first one was pointed out to me by Vaigret, in as much as selections could be made in such a way that people could guarantee themselves a fairly good score despite picking several winners.
This information was gratefully received. (I will always listen to people's opinions when there is clear merit supporting their viewpoint)
I don't think anyone is actually exploiting this, but nevertheless it pointed out that the penalty for picking winners is not severe enough. After all that "should" be the main aim of the removals process and the main aim of each removal person.
This is actually fairly simple to resolve. I have discussed this with Vaigret and we feel that a "doubling" of the negative points for winners should do the trick. (The positive points for losers will remain as it is now)
So we will implement this in 2018 only.
The other "flaw", which I have always been aware of, is as follows. (Those not particularly mathematically minded may want to switch of their sets) :=)
1) When people do selections at some meetings and not at others a disparity of "fairness" could be created.
2) Say for example "REMOVER'S A, B, C, D and E" participate in the first GO-LIVE of 2018 and there are some horrendous results causing the average score to be 10pts and the average number of winners 3
3) Then on the next GO-LIVE, (REMOVERS A, B, D, F and G) participate. In this one all the fancied horses win and the scores are very high. Lets say and average of 1 winner per person and average points of 40
4) Clearly REMOVERS F and G are severely advantaged, while removers C and E severely disadvantaged, whilst A, B and D in the middle
I always hoped this type of "stuff" would even itself out, but I don't think we have enough "volume" in a year for that to be the case.
OK, so what is the solution?
In planet utopia, all removers would participate in all GO-LIVE's, but clearly this is unrealistic.
My thoughts on it are as follows:
We could adjust people's scores so that it is in relation to the average score on that day. So lets say each of the above two days there were 20 selections made and, for simplicity sake, no NR's
This would mean that the average score on the 1st GO-LIVE was 0.5pts per selections and on the 2nd GO-LIVE an average of 2pts per selection.
We could then base people's scores around the average of that day. So say someone got a 0.7 average on the first day, their score would be adjusted to a +0.2 average, if one of their opponents got a 0.1 average, their points would be adjusted to a -0.4 average.
If some one had a 1.5 average on the second GO-LIVE that would count as -0.5 and if some had an average on 2.8 on the 2nd GO-LIVE that would count as +0.8 and so on.
Clearly this makes the scoring more complex/but fairer, the problem then comes along when there are only a few removers on any given day. There may not be enough removers on the day in question to give a "fair" average score. (What do we do if only 1 or 2 removers?)
Anyhow, I know the likes of Vaigret, has good thoughts on these type of things, and I would encourage others to do the same.
So please post on here with any potential solutions. The "fairer" the process is in 2018 the better the system will work for all. Now is a good time to discuss options, so hopefully we can reduce "flaws" come the start of 2018.
I appreciate that added complexity to an already "fairly complex" scoring system is not ideal, and may put people off, but we could potentially get used to it, as fairness and a system that is fit for purpose, has to be the main aim imo.
Clearly, by the nature of things, we will never get a "perfectly fair" system, but we can strive to make it as fair and fit for purpose as possible.
Please post your thoughts.
Cheers,
G
There are a few flaws in the scoring system for the removers system and being a poker player, I am always looking to sort out leaks. :=)
The first one was pointed out to me by Vaigret, in as much as selections could be made in such a way that people could guarantee themselves a fairly good score despite picking several winners.
This information was gratefully received. (I will always listen to people's opinions when there is clear merit supporting their viewpoint)
I don't think anyone is actually exploiting this, but nevertheless it pointed out that the penalty for picking winners is not severe enough. After all that "should" be the main aim of the removals process and the main aim of each removal person.
This is actually fairly simple to resolve. I have discussed this with Vaigret and we feel that a "doubling" of the negative points for winners should do the trick. (The positive points for losers will remain as it is now)
So we will implement this in 2018 only.
The other "flaw", which I have always been aware of, is as follows. (Those not particularly mathematically minded may want to switch of their sets) :=)
1) When people do selections at some meetings and not at others a disparity of "fairness" could be created.
2) Say for example "REMOVER'S A, B, C, D and E" participate in the first GO-LIVE of 2018 and there are some horrendous results causing the average score to be 10pts and the average number of winners 3
3) Then on the next GO-LIVE, (REMOVERS A, B, D, F and G) participate. In this one all the fancied horses win and the scores are very high. Lets say and average of 1 winner per person and average points of 40
4) Clearly REMOVERS F and G are severely advantaged, while removers C and E severely disadvantaged, whilst A, B and D in the middle
I always hoped this type of "stuff" would even itself out, but I don't think we have enough "volume" in a year for that to be the case.
OK, so what is the solution?
In planet utopia, all removers would participate in all GO-LIVE's, but clearly this is unrealistic.
My thoughts on it are as follows:
We could adjust people's scores so that it is in relation to the average score on that day. So lets say each of the above two days there were 20 selections made and, for simplicity sake, no NR's
This would mean that the average score on the 1st GO-LIVE was 0.5pts per selections and on the 2nd GO-LIVE an average of 2pts per selection.
We could then base people's scores around the average of that day. So say someone got a 0.7 average on the first day, their score would be adjusted to a +0.2 average, if one of their opponents got a 0.1 average, their points would be adjusted to a -0.4 average.
If some one had a 1.5 average on the second GO-LIVE that would count as -0.5 and if some had an average on 2.8 on the 2nd GO-LIVE that would count as +0.8 and so on.
Clearly this makes the scoring more complex/but fairer, the problem then comes along when there are only a few removers on any given day. There may not be enough removers on the day in question to give a "fair" average score. (What do we do if only 1 or 2 removers?)
Anyhow, I know the likes of Vaigret, has good thoughts on these type of things, and I would encourage others to do the same.
So please post on here with any potential solutions. The "fairer" the process is in 2018 the better the system will work for all. Now is a good time to discuss options, so hopefully we can reduce "flaws" come the start of 2018.
I appreciate that added complexity to an already "fairly complex" scoring system is not ideal, and may put people off, but we could potentially get used to it, as fairness and a system that is fit for purpose, has to be the main aim imo.
Clearly, by the nature of things, we will never get a "perfectly fair" system, but we can strive to make it as fair and fit for purpose as possible.
Please post your thoughts.
Cheers,
G
Comments
As regards your point re some days being fairer than others I think it will work out over the year especially as the removers are a small band anyway. I wouldnt complicate it anymore to try and get complete parity and give extra working out as well.
I think there was some good scores when I missed out the other day but thats not to say I would have got a high score as well if i had participated and anyway I didnt go down in average so didnt lose out.
V
happy to go with anything you change