Sunday, September 28, 2008

USCL Top Board Honors, Week 5

Here are the Top Board Honors, for Week 5 in the United States Chess League.

Note that players must have played at least TWO games in the 2008 season to be eligible for Top Board Honors. Rankings are done by USCL rating.





Board 1


1 2617 Jaan Ehlvest Tennessee
2 2609 Sergey Erenburg Baltimore
3 2606 Hikaru Nakamura Seattle




Board 2


1 2593 Josh Friedel San Francisco
2 2538 Drasko Boskovic Dallas
3 2537 Eli Vovsha Queens




Board 3


1 2461 Sam Shankland San Francisco
2 2460 Gregory Braylovsky New York
3 2454 Mehmed Pasalic Chicago




Board 4


1 2571 Marc Esserman Boston
2 2364 Eric Rodriguez Miami
3 2350 Angelo Young Chicago

23 comments:

Greg Shahade said...

Are you sure Lenderman and his 4-0 record hasn't made the list?

Ilya said...

no Jorge either...

Bionic Lime said...

Lenderman's USCL rating is 2488. He has played two games on Board 2 and two games on Board 3. Most recently, he played on Board 2, which makes him eligible for Board 2 honors only.

Jorge has played two games on Board 1 and one game on Board 2. That makes him eligible for Board 1 honors only, and his USCL rating is 2581.

Arun Sharma said...

You listed a rule last week as:

2. Players must have played the majority of their games on that particular board. If they have played the same number of games on different boards, the most recent board on which they played is the one for which they are eligible for the Top Board Honors.



Does that apply for majority this season or majority overall? Just struck me as odd to list Boskovic amongst the Board Two players considering he's played only one match on Board Two, and the rest (like twelves games I think) were all on Board 1. If your system judges the eligibility of a player for a particular board by only this season's games, then that's correct, he should be Board Two. But that does strike me as a bit unusual given the disparity in the amount he's played on each Board total, to list him as a Board Two really based on only game there, and likewise Jorge not being listed at all since he happens to have played two games on Board One this year and only one on Board Two, but counting all his games, he's clearly more often a Board Two.

I do see why using the your system might be more reasonable in many cases, given someone like Shankland would still be listed as a Board Four player if you count all of his games which, given his current rating, seems ridiculous. Honestly, I'm not sure what the best way of doing this to balance those inequities would really be...

Either way interesting system for sure, I too am surprised though by some of the names which are and aren't on the lists.

Ilya said...

Seems to me this system sucks and isnt worth being linked to on uscl website, how can best performers in the league be ineligible for top board honors because of some technicalities...

Bionic Lime said...

It is for this season only.

If you don't like the system, my suggestion is that you either (1) ignore it, or (2) suggest a different way of doing it.

Ilya said...

I did suggest something i.e. to ignore it by not linking to it. You have to be first one to admit your system sucks if it omits best perfoming players...it is simply unaccetable, either place them by the highest board they play on, by the board they have most points on, by lowest board, however you like--thats your problem.

Bionic Lime said...

For what its worth, Ehlvest (2.0 - 0.0) and Nakamura (2.0 - 0.0) are better performing than Jorge, and Erenburg (2.5 - 0.5) is tied with him.

Lenderman would be the top performing Board 3, if the criteria were such that he was slated for Board 3.

That being said, I did this Top Board Honors thing as a lark, but now I've put a little more thought into it. For next week, instead of actual ratings, I'll be using a modified performance rating, which actually makes a lot more sense.

Greg Shahade said...

Why not use the USCL rules for all star ranking the boards? For us, you can be eligible only for the board which you have played the most games at during the season. If you have played on two boards evenly, then you are eligible for either of them.

Seems like it'd be most logical to base it on our system, since this is the kind of thing people might care about in terms of all star teams, however if you are going to use your own independant system, then it really has no relevance to our all star teams at the end of the season.

Greg Shahade said...

arun obviously if someone has played two boards evenly, they should be eligible for either one. There's no logic to "whichever board they played most recently". It's just a completely arbitrary way to go about it, and doesn't follow USCL procedures at all.

However Jorge should be considered a Board 1 player, as I believe he has played the majority of his games there this season, and all three of the players listed have had tremendous results.

Bionic Lime said...

Greg, it is totally up to you to use whatever I come up with (power rankings, rating system, top board honors). My systems are based on my experience doing modeling and forecasting.

Greg Shahade said...

Well I don't use your work, I just link to it for USCL fans to read. I'm just saying that it goes against the standards that we use for determining all star prizes.

For instance if someone plays the first 6 matches on board 3, and the final match on board 7, your system would have them listed as a 2nd board, which they obviously shouldn't be eligible for, in regards to any type of all star teams.

Greg Shahade said...

Also in regards to Ilya's comments:

I don't believe that these ratings incorporate a USCL players entire career. So even if someone is doing exceptionally well this season, they may not be on the list.

You can't deny that both Ehlvest and Erenburg have had great records in the USCL. I'm a bit surprised that Nakamura's rating is so high, but I suppose he has played very strong competition.

Sammour's rating in bionic's system is probably hurt because he played so many guys on Board 2 last year, like Bryan Smith, Tegshsuren Enkhbat, Jake Kleiman and etc, all of whom probably have much lower ratings than the guys that Erenburg, Ehlvest and Nakamura have played against.



Also we will publish pretty much everything about the USCL as long as it's not offensive. There is no reason not to publish this, as it's simply a list of the highest rated players at each board throughout their USCL career, which cleary has some relevance.

Greg Shahade said...

Also bionic, why is Jorge not listed for Board 2, while Boskovic, who has also played only one game on Board 2, is listed? Seems like an inconsistency.

Arun Sharma said...
This comment has been removed by the author.
Arun Sharma said...

Greg several of your points/posts make no sense to me, but maybe you can elaborate.

"arun obviously if someone has played two boards evenly, they should be eligible for either one. There's no logic to "whichever board they played most recently". It's just a completely arbitrary way to go about it, and doesn't follow USCL procedures at all."

When did I suggest using most recent board as a way to settle when someone has played evenly? This rating system is the one using that tiebreaker, and I never suggested using it, I was merely quoting a rule Bioniclime had mentioned in his last week's post.


"However Jorge should be considered a Board 1 player, as I believe he has played the majority of his games there this season, and all three of the players listed have had tremendous results. Also bionic, why is Jorge not listed for Board 2, while Boskovic, who has also played only one game on Board 2, is listed? Seems like an inconsistency."

Didn't you answer the latter question with your first statement? Seems perfectly logical to me when using the aforementioned system why Boskovic would be listed as a Board Two and why Jorge would be listed as Board One.


"I don't believe that these ratings incorporate a USCL players entire career. So even if someone is doing exceptionally well this season, they may not be on the list."

I think you just contradicted yourself again here. If the ratings were ONLY for this season, then of course someone doing exceptionally well THIS season would be MORE likely to be on the list.


"Sammour's rating in bionic's system is probably hurt because he played so many guys on Board 2 last year"

You just mentioned above (in the same post!) that "I don't believe that these ratings incorporate a USCL players entire career." If that's the case, then why would Sammour's rating be hurt on account of the people he played last year??



Ok, enough Greg bashing (for now!). It does makes sense to why Lenderman and Jorge are not listed; Lenderman by the rules is only eligible for Board Two and even though he is 4 - 0, his performance rating would generally be lower than someone who simply went 2 - 0 on Board Two (as Friedel essentially did or close). Likewise for Jorge, even though he has a similar record to Ehlvest, Erenburg, etc., he did play one game on Board Two, and even though he won there, that effectively counts against him since everyone is undefeated or close (and obviously when everyone is undefeated winning on Board Two is a detriment to one's performance rating when comparing to people who have won and played only on Board One). Effectively, for both of them, their extra wins (Jorge's on Board Two and Lenderman's on Board Three), effectively were a cause of them not being on these lists which clearly doesn't strike me as being fair.


In any case, these seeming discrepancies of certain exceptional players not being on the lists do make sense if you consider the rules of the system, but given the results, I have the same apprehensions as others. Lenderman, whichever Board he's listed at, being 4 - 0, not being a top board honor is just ridiculous to me. With Jorge, that makes a bit more sense, as the Board One's who are listed instead have all performed similarly to him and him being below them doesn't seem ridiculous. But for instance listing Boskovic who's record is 0 - 0 - 2 this season on Board Two, rather than either of them who are 4 - 0 and 2 - 0 - 1 just doesn't add up to me in something you title as "Top Board Honors" even if by the system's rules it is technically correct.

Well in any case, that's my view of things. I look forward to seeing what your new system next week comes up with!

Bionic Lime said...

The current system looks at current USCL rating (which is a lifetime thing). The board they are put on is based on this season only.

I agree that there's a much better way of doing this, and that's what I'll incorporate next week.

Greg, one thing you stated was incorrect... A person is eligible for the board on which they played the MOST number of games this season. The only time you use most recent is if there are two (or more) boards on which they played the same number of games.

Greg Shahade said...

arun i made a ton of typos. I meant that I DO BELIEVE they incorporate the entire career.

I also was agreeing with you on the first point.

Also after my first post I realized something was fishy because Sammour and Boskovic should both count as board 2 based on that format, so that's why I made another post.

Greg Shahade said...

ahh now I get it bioniclime. Im dumb.

Greg Shahade said...

Just curious but did Esserman start with an artificially high rating because of his strange emergency placement on Bd 1 last season, or am I mistaken about the format?

Arun Sharma said...

"arun i made a ton of typos. I meant that I DO BELIEVE they incorporate the entire career."

Well that makes more sense then, but in regards to the people we tend to be discussing, it doesn't really fit exactly. If someone is happening to do well this season (like Lenderman) who hasn't played in the League previously, it is to such players' advantage for these stats to include ALL games since it's much easier to have a record of 4 - 0 or something similar over a small sample size. Of course, someone who was mediocre last year, who's doing amazingly this year, that point could well be valid for.

In general, the system of deciding which board a person is eligible for if they happen to have played evenly on two boards by which board they happened to play last, doesn't seem like a bad one, but I still wouldn't want to use it for when we decide All Stars honestly. If Lenderman continues to perform as he has and is still eligible for either board by season's end, it seems too arbitrary to decide which board he should be an All Star for by that criteria (something that which obviously would affect other people's positions on the All Star Team, which doesn't seem right).


"ahh now I get it bioniclime. Im dumb."


Well, at least you finally came up with something that I don't have to point out all the mistakes in (you know, because there aren't any)!

Bionic Lime said...

I've come up with an alternate system, which I think makes a lot more sense than this one. I'll post it tonight or tomorrow.

As a preview, Lenderman will show up as top player on Board 2.

Greg Shahade said...

Also note that Im pretty sure someone starts with a lower/higher rating depending on which board is the first they played on. This seems a little odd, especially if the first game they played is not on their natural board due to some fluke. You'd imagine that this would drastically affect their rating for many games. The most obvious example of this is Esserman, but also if Lenderman played Board 2 in his first match, I suspect he'd be much higher as well.

Am I correct (when I post here it seems I'm usually not correct)? If so this seems a bit strange.