For a 28 year old + 2100 to get 3 IM norms in a year.....it's not 1000 to 1 but 10 to 1 is safe IMO. Maybe I'm wrong, I guess is someone had the incentive to quit their jobs and social life, hire a G.M trainer and travel around for a year spending 10 hours a day studying chess, who knows what could happen.
Fruitless discussion though, prop bet is too boring and hard to set up IMO.

To clarify, I meant that I would lay 1000:1 that a 2136 USCF who is not underrated could not become an IM in one year, not that they couldn't get three IM norms. Dire is ineligible because he is probably already NM strength.

This is actually a fairly easy task to accomplish if its just the rating you want.
Just lose 50-100points on purpose, so that youre a bit low rated.
Then the next rating period you play tournaments like hell.
This way you may actually get a rating far far higher than 2250.
Only drawback is that youll lose most of it during the third rating period...

The practical problems mentioned in the thread are unfortunate, but true...

Also, going from 2100 to IM in one year is really lol for a tournament veteran imo.

I just lost 15k today playing poker, so I have to make sure I don't go busto anyway.

I think there are a lot of people who'd take the FM bet for a decent amount if there was a way to get around the technical difficulties.

Wat

You have no idea what you're talking about, the USCF rating system doesn't work this way at all.

Please post which limits you were playing and Hands where you were either coolered/pwned.

and GL in recouping the losses.

100/200 limit holdem on Stars, hand I remember most is TT<44 on like 85224 board against a solid reg who raise/called turn, I'm still too tilted about this to post HHs, but thanks

He posted his PS screen name in the OP. http://www.pokertableratings.com/sta...rch/drarpadelo

GL DrElo.

I've played with Elo a few hands on 10/20 back then before i moved first to FTP and then to mixed games. Definitely a very good limit player.

Ok, explain where Im wrong then.
Pretty much every other rating system can be manipulated very easily in some way and the USCF is no exception.

I'm having a little trouble understanding how your plan might work. Let me try to sum up and you tell us if this is correct.

Step 1: Lose rating points so that ... A) you play in a lower section? (I'm not sure what you're going for here. Is this right?) Or B) lose rating points so that when you perform "normally" again, then you gain more rating points than you would have?

A) would work horribly since you'll gain far fewer points in a lower section. This would work if you really wanted to sandbag in order to win money.

B) Won't work because each tournament is rated sequentially, regardless of when the rating period ends. This means you cannot overshoot your true strength by sandbagging then playing normally.

For example: Your true rating is 2200. You intentionally donk off 200 points to a new rating of 2000. Now you play for real and gain 40 points in a single tournament. You won't keep gaining 40 points every tournament until the end of the rating period. Your 2nd tournament's rating change will be based on an initial rating of 2040. Say you gain 35, then your next tournament the rating change will be based on 2075, etc, etc. So you'll get back to 2200 and pretty much stop.

Step 2: ????

Step 3: Profit!

Also he has some strange concept of "rating periods".

Its pretty meaningless to discuss these things with you since you dont seem to have even basic understanding how the rating system works.

Two 2000 plays each other. The player who wins will get about 15points.Nothing for draws.

If a 1900 or even an 1800 rated player beats a 2000 he will get 20-25points plus a lot of points for every draw.
Since the USCF ratings do not get adjusted after every single game like on internet, its fairly easy to see how this can be abused.

There should be are many ways around this obstacle too if you really want to.
You can for example arrange 20round tournaments etc.

Alright, I'm just gonna put you on ignore since you have no clue what you're talking about and don't seem to care to learn. Ratings are calculated based on your actual tournament score compared to your expected score based on initial rating. Only the tournament as a whole matters. The expected score varies from tournament to tournemnt because rating is calculated at teh end of every tournament. Yes, obv you could have a 20 round tournament and you could simply pay various high rated players to lose to you. That's not the point.

Swingdoc, I was just thinking that I felt bad for you when I saw you trying to reason with him .

Pretty much whatever you say, any rating system can be abused.
Even on the Elo list there has been numerous 2600 players whos actual strength has been more like 2200-2300.
A friend of mine made a good first tournament and entered the Elo list with 2350. His actual playing strength was never higher than 2050-2200.

I dont live in the US, so I cant say Ive studied the USCF ratings too carefully, but I would be highly surprised if this particular one rating would be more accurate and have less flaws than any other rating system.

Swingdoc,

actually Fide ratings are based on the actual games played, not the tournament as a whole. They changed this a few years ago. Your actual rating only changes after the tournament is finished (that's probably what you meant), but it is calculated based on the single games.

And also, with Fide rating, a scenario like described could be possible because they only publish rating lists every two months.

A little thought experiment:
Say your playing strength is 2100 Fide Elo, but you are for some reason only rated 2000. Your goal is to profit from the "wrong" rating and reach FM (2300) in a single rating period (which is possible because the rating is only then updated).

So you would have to play 300/15 = 20 points above your expectation (15 being Fide's K-Factor). Assuming your average opponent is rated 2100, your actual score will be 0.5, but your expected score will only be 0.36 per game (as you are rated 2000). So you win 0.14 points per game on average which translates to 0.14 * 15 = 2.1 rating points. So you only have to play 300/2.1 = 143 games, which is roughly 2.5 games a day every day for two months. Nothing easier than that... (which means just impossible).

Sorry for the long post, I figured this out some days ago and don't think it is worth a post, but I couldn't resist.

Of course, I don't have the slightest idea what K-Factors and rating periods the USCF is using, so all this is probably not valid there.

Edit:
I know the calculation is partly wrong because I didn't take into account that the point distribution is not linear but standard normal, and there would be different results for different opponents' ratings. But who cares...

There is some truth amongst the rambling above, rating systems are indeed abusable. The Crisan case and the Myanmar case come to mind. There was a hilarious game between Mohr and Crisan from the Portoroz tournament (mentioned in the article above) where Mr. Crisan (2635) lost a rook ending with the starting position being somewhat equivalent of w: Kd4 Ra4 pe4 b: Ke6 Rb8.

The K-factor is 10 for very experienced player, 15 for most players maybe.
Some other with less experience got 25.
Like you mention, its probably fairly meaningless to attempt something like that if youre an experienced player with K-factor 15 or less.

A very important factor youre missing here is that the 2000 rated player who play an equilly good opponent, but rated much lower will press much harder for the win unjustified and accordingly suffer many more losses than he should do.

Anyway, the prop bet OP talks about includes moving from 2128 to 2250.
Only a complete idiot who has absolutely no understanding at all about how the rating system works would take a bet like this.

A very important factor you are missing here is that this does not matter at all as we are speaking about expected values and neglecting variance. A poker player should know the concept.

You basically say that the 2100 player plays weaker when he plays against a lower rated player. This is illogical and irrelevant in mathematical terms. A 2100 player will play 2100 against better and 2100 against worse players, variance excluded. If the 2000 player is in fact 100 points underrated he will make 50% in the long run.

Also, the prop bet assumed first losing about 100 rating points, so it is not about going from 2128 to 2250, but from something like 2020 to 2250. If he doesn't lose the points first it is practically impossible for a correctly rated 2128 to gain 120 points during two months (assuming K being 15, as 25 and 10 are irrelevant for the players we are talking about). And my rough calculation from above shows that it is almost as impossible to gain 220 points even when 100 points underrated.

Ajezz, good point. I should have qualified in my post that I was talking about USCF ratings specifically, which is what the initial prop bet was referring to. Those most definitely are calculated based on tournament performance and not game performance. Thank you for clearing up what may have been confusing some

That wasnt what I said... I said that the higher rated players usually takes more risks (Perfectly justified if the opponent really is weaker like the rating suggests!) in order to play for the win.
You cant play the exchange slave, exchange french, drawing lines in the caro-kann and let your much weaker opponent make easy draws.
Chess doesnt work that way.

The USCF rating system awards bonuses when you have a super high performance rating in a tournament relative to your current rating, so if you lost like 200 points you could get some massive bonuses, but even then I'm not sure if it compensates/is worth it.

Plus everyone would know you're a sandbagger, you'd get banned from USCF tournaments, etc, etc.

Wouldn't be worth it for 10k imo.

Hmm, $10k to never play a uscf rated game again...

Somewhat tempting in theory.