Chess as a fantasy sport

Those of us who have played fantasy sports know that it is generally a prediction game: we pick players who we believe will record good stats in an upcoming game, and add them to our squad, hoping that they will deliver and win us more points than the players that out opponents picked. So, what if we we wanted to play a fantasy sports game involving chess players? That is exactly what I have in mind, and am planning to launch as soon as possible - a fantasy league where we predict the elo adjustments of top chess players! The rules that I have in mind thus far are:

1. To participate, one must pick 10 players from the Jun. 2017 top 100 list by Jun. 31.

2. On Jan. 1, 2018, the fantasy scores of all players that have been picked will be calculated. The person with the highest average fantasy score among all of his players wins.

A player's fantasy score (FS) is dependent on his rating on the Jul. 1, 2017 (R1) and Jan. 1, 2018 (R2) official rating lists. The formula for calculating fantasy scores is:

FS = R2 + 2*(R2-R1)

For example, if a player is rated 2750 in July and 2770 in January, his FS is 2770+2*20=2810. If someone else is 2790 in July and 2760 in January, his FS is 2760-2*30=2700.

To maximize their scores, people will need to find a balance between choosing players with an already high elo, and those who they think are likely to increase their rating in the coming months.

These are all of the rules I have thought out in detail thus far, but I think it would make the game more fun to add a means of trading players - i.e. exchanging the players one selected earlier for others that one did not select. I have some thoughts on how such a mechanism could be implemented, which I will lay out in detail in a future post. In the meantime... who is interested in participating in this fantasy league? And are there any suggestions on how the game can be improved?

Just choose the currently highest rated player(s). Your formula is flawed.

Rethink it.

I could suggest a fix, but it would be crtiticised.

But the present one is wrong

But the idea is interesting, to say the least.

First of all, you're choosing 10 players, not 1. Most people will probably pick Magnus Carlsen as one of their players, but there are still 9 others to choose.

Also, choosing Carlsen might not be the best idea. If he starts with 2830 and ends with 2830, he will have a lower FS than someone that goes from 2780 to 2800. And if he goes down, then there will probably be many players with a higher FS than him.

That's not to say that the formula is infallible, of course. If you would like to propose something else, please do. One idea I have is that if it really is too biased toward players currently with a high rating, a simple fix might be to change the adjustment multiplier to 3 - in other words

FS = R2 + 3*(R2-R1)

But maybe that would be too biased toward players under 2700, who have a lot more room for improvement?

I just thought of another flaw. The comp would be open to rigging by entrants who conspired to do so.

Perhaps the candidates would have to have characteristics which made rigging unlikely. 

How would that work?

Say I choose yourself. Before july 1, you sandbag madly.

That it how it would work.

Please read #2. I had a typo. I forgot the 's' in players. I got the choose 10 player rule.

My typing let me down.

What is new?

Eligible players are those in the world top 100 in the Jun. 2017 official rating list.

I just noticed the typo in my original comment (where it said top 10 instead of top 100), and it has been fixed.

See my edit to #3.

Among the top 10 players, I would not choose anyone who I think will drop in rating in the next 6 months - e.g. Mamedyarov. If he drops from 2800 to 2770 that would absolutely kill my score, especially if the multipier is increased.

I'm in! 

How do we play?

I'd like to begin as soon as there is a decent amount of interest and rules are settled. In the meantime, do you have any thoughts on the rules I have proposed?

I understood nothing till post #10.

It is called fine tuning, and even though he might not admit it, I think Macer appreciated my input.

You would probably have whinged if we had not done the bits you could not understand.

I didn't understand that either.

O.K.

I hope you understand THAT!

At least part of it.

I do understand YOU though, and the other trolls around the place.

H--- s---! Did you just call me a troll?

Yes! Yes! Yes! This is the best day of my life! The moment I have been waiting for! I have officially became a troll!

It becomes official only when you are initiated!

I surmise only, there. But, they will verify it, or otherwise. But, the talent scouts are out and about.

I will admit it. I'm looking for input right now - that and interest in the project.

Anyway, a TL;DR version of everything before #10:

I want to create a fantasy chess league where people compete in predicting the rating adjustments of elite gms, similar to how people predict the stats of players in fantasy sports. The precise rules are still being fine tuned, as you put it, but as I envision it currently, both the actual ratings and the rating adjustments of the players will play a role in determining their fantasy score, which is the score that people playing the fantasy game want to maximize.

I have proposed a formula for the score in the original post, and the topic of discussion is whether there might be a better formula.

Come on! An initiation?

Thank you. Got it now. So what I have to do is predict the rating of players every once in a while?