Reforming the NM title to avoid corrupt gaming of the system

All Things Chess
1

Greetings! I’ve heard a lot about the problem of people “gaming the system” by exploiting one’s TD status to set up mini-matches or organizing events with their friends and/or colleagues to help someone to unfairly obtain the National Master title. I hope to prevent this from happening in the future. Maybe you’re aware of similar shenanigans or have some valuable insights on the topic, which I would greatly appreciate in seeking change to preserve the competitiveness of the game, not to mention the integrity of USCF and one of the highest titles that it awards.

My intent is not to target any one person specifically for alleged infractions here, but rather to take a more productive approach and to help establish rules that will aim to prevent this from happening in the future. From what I understand at this point, it seems that an ADM (advanced delegate motion) is the way to do so. Or please let me know if there’s a better way. I’m hoping that those with a deep understanding of the inner workings of US Chess / the titling system could offer their two cents on the matter.

To further clarify my intent before getting into the thick of it, while I’ve obtained the NM title myself, I don’t seek to take action here in order to “jealously guard the title” or something like that. From the standpoint of seeking fairness and the best for USCF members (and the organization itself), I do think that the title should be viewed as something that one really needs to put in the hard work to achieve – not another area for cheats to seek self-aggrandizement and spoil it for aspiring players who want to win the title in an ethical manner.

And since I mention self-aggrandizement, I think the problem of corruptly obtaining the title may extend beyond the “title winner” themself. For example, with a small group – such as colleagues of an affiliate that I referenced above – the entire group may benefit in terms of raising the profile of their organization, allowing it to earn more profit in the form of camps, simuls, classes, and just generally drawing folks in based on the widely understood legitimacy of the (ill-gotten) title.

I found it interesting that literally every time I’ve mentioned this issue to someone recently, they’ve said, “I know someone who has done this” (so common that one said the individual they knew was present at the very same event)! So it seems like there are a lot of people who want to make some change here to ensure that the title can’t effectively be gamed – and movement on this front would seem to align with anti-cheating measures that are being taken more broadly.

I apologize for what may have been a little rambling there; just wanted to present some context. As for the specifics, I’m not an expert on how the ratings and titled systems work, so I hope that I can obtain some input from those in the know. Some questions that come to mind:

  1. Are there any efforts currently underway to more strictly specify how the NM title is to be obtained?

  2. I see that, while the candidate master title (hopefully not to be confused with FIDE CM…that’s another, more petty matter) is “norms based,” it’s not clear that the NM title is as well. Or is it? While I see that I have 5 C’s and then 3 M’s in the events leading up to NM, I saw another who obtained the NM title with only four C’s and zero M’s (I won’t link to this example because, again, my intent here isn’t to shame any individuals, but to avoid what I see as corrupt and unfair behavior from occurring more broadly in the future). Also, when I look at the player profiles of NMs, I see that they have “National Master, Candidate Master (norms based)” listed, which appears to only apply to CM and not their NM title listed before it (unless the parentheses would ambiguously apply to all prior titles listed). If three strong performances (what is the standard, 2300 performance?) are in fact required to obtain the NM title, then “gaining” the title by shamelessly beating up a bunch of lower rated opponents would largely, perhaps, be precluded, in which case fewer changes would likely be needed. Which is why I want to nail down on the specifics here. My impression has been that NM titles are simply based on reaching 2200 and not norms based, though someone stated otherwise so I’d like to clarify this. If someone can provide clarification in this area it would be greatly appreciated. Beyond this, even if we’re to assume that the NM title requires the performance based norms in addition to reaching 2200, just being able to rack up points incrementally in small arranged matches (I hear of “the Borises” on this topic) is a problem in and of itself, since one can claim they’re effectively “master level” and corruptly exploit the various benefits (including financial) that the title confers.

  3. I see no mention of titles in the USCF bylaws. There is a 2016 memo on titles, though it seems only life master and senior master titles are mentioned but not national master. Is there an updated memo? If new language were to be proposed to further stipulate how the NM title were to be attained, where would this language go and what would it state?

  4. Does anyone have any ideas, maybe based on foul play you’ve witnessed, for how to avoid such corrupt arrangements? How about training (and possible penalties) for TDs who attempt to make such corrupt arrangements, i.e., prosecuting the procurer? Avoiding small matches / “insider points-trading”? Open tournament requirement to weed out those who simply play in small rings? Performance ratings (again, if not already in effect)?

My goal would be to make this as simple as possible so as to not confuse people about how the title is to be attained, all while ensuring that one must indeed perform at a high level in order to get the master title. Individuals would therefore be disincentivized from seeking titles in corrupt ways and the ethics committee would have, hopefully, an easier job, assuming they get a decent number of complaints on this topic already. This would seem to refute the argument that the best solution for these matters is for the ethics (or most appropriate) committee to deal with these arrangements on a case-by-case basis. Instead, the best solution ought to be the one that is most efficient, clear, discourages such behavior, and prevents it from happening so that these ethics violations will be far less likely to occur in the first place.

As you can gather, I’m operating on some assumptions and want to know what I don’t know, so if you could fill me in and maybe propose some ideas of your own for reform I’d much appreciate it. Hopefully something can happen at the delegates meeting this year to clarify matters and to prevent this form of gaming the system to occur in the future. It really doesn’t matter to me who does it, but just that some action is taken if the will is there – which, as I suggested above, I think it is. I’m willing to help however I can.

Thank you,

David Bennett

–

David Bennett

Chess Coach, National Master

Black Knights & DC Chess League Board Member

www.chessprofessor.net

2

The NM (National Master) title is awarded strictly by a post-tournament rating of 2200 or higher. The LM (Life Master) title is based on norms. The NM and OLM (Original Life Master) titles far predate the “norms” system and some oldtimers don’t think highly of that innovation. That said, I earned all three USCF titles over the span of less than three years (in the above mentioned order).

Sadly, I narrowly missed out on FM (FIDE Master), but earned FT (FIDE Trainer) instead. Hence I sign this post as…

NM LM OLM FT Michel Aigner

1 Like
3

Thank you for your clarification that NM is not a norms based title @fpawn. That’s super helpful. And congrats on your cocktail of titles!

Now the question becomes: would requiring performance-based norms for NM in addition to hitting 2200 be an effective catch-all to largely prevent such attempts at “gaming the system” to occur? That is, would a simple change to how the NM title is gained solve the woes that I mentioned in the post? (By the way, I don’t even think for the sake of this reform that one would have to delve into the nature of other master level titles, since the NM is the first one and for most members I think the main one, with the others being cherries on top, no? Especially if we’re talking about public perception and “emoluments” for the title holder.) Or would more targeted language also be needed, such as no random small matches, etc.? Although I imagine a norm couldn’t be obtained through a 1 on 1 match…? At any rate, the rule would need to parsimoniously root out all sort of chicanery imaginable.

This brings to mind something that I had intended to mention in the realm of “big city elitism.” I could imagine one arguing that they have these small, arranged matches to get to 2200 because they live in a small town (big fish in a small pond) and don’t have access to the big clubs or tournaments. (The argument I have heard is that they just liked to play solely in their own organization’s events.) Somehow I don’t buy it, though. After all, it is the “national master” title, so one would, I think, be expected to compete to some degree nationally. And if this were a burden, then more systemic issues could be addressed, and donors/charities (including the U.S. Chess Trust) would perhaps be able to provide some assistance so that the individual vying for the title would in fact be able to compete in larger events and face the stiff competition that others who attained the title had to encounter.

4

As far as TDs gaming the system goes, such gaming in the past had resulted in retraction of the NM titles, correction of the ratings and suspensions applied to the TDs. So I also know of people that have done it, but they have not gotten away with it.

Currently the simple NM title is as purely ratings based as FIDE’s CM and FM titles. It can remain that way. There are already norm-based titles in the US Chess system (you can look at your own Member Services Area page to see your titles). One of the reasons they were created was to have performance based titles people could point to, but hardly anybody pays attention to them.

As far as the big city elitism goes, being in an area with a lot of chess makes it easier to become a National Master (or National Tournament Director [NTD] for another example). Travel to farther away tournaments does help get NM and is pretty much required for NTD (if you have no intention of traveling to areas with large events then there is no real need to be an NTD).

2 Likes
5

Interesting, I am a new TD, and the only TD in Alaska. We only have 100 USCF players in the state. It costs over $1000 to fly anywhere to play in a tournament, not counting hotel, food, gas/vehicle/rides.

I have no problem with your comments, but whatever you propose should not limit people in those small towns from getting a deserved title who are earnestly seeking it and not gaming the system.

I can’t afford to fly out of state to play any events. So I run small quads every other month to try to get myself an established rating, just so I can step up to local TD. (No danger of me being a NM - :grinning: ) so I am sure there are some cheaters/gamers, but I can see how honest folks in small towns could just be doing the best they can with what limited resources they have got.

2 Likes
6

Some years ago I floated a suggestion that in order to cross a 200 point ratings threshold, a player had to actually compete against players who were one class below that rating or higher. In other words, in order to become a 2000 player one had to have a game in that event against at least an 1800 player. (The Ratings Committee thought the proposal was too complicated.)

At the time we had a number of examples of players who crossed a threshold (usually expert or master) by playing only players at least 200 points below that level and creeping up a point or two at a time.

The ratings system was tweaked about 10 years ago to keep ratings in floating point rather than integers, in part to limit ratings creep. I don’t know if ratings creep titles are still as big a problem as they were before that.

Note that the Candidate Master and higher norms-based titles have both a norms requirement and a peak ratings requirement. CM or LM norms are not easy to get, a player has to have a performance more than a full point above expected performance to earn a norm. I’ve seen many events where players had a rating gain of 20 or more points but didn’t earn a norm.

Titles and ratings are not part of the rules or Bylaws, so this is technically no longer a matter under the Delegates’ authority, that’s now the EB’s responsibility.

1 Like
7

Could you be a bit more specific about what you see as “gaming” the system? Matches are subject to rules, the most important of which from a “gaming” standpoint is that a floored player consents to lowering their floor. If a disguised match is detected, match rules can/will be applied. We no longer round positive results to the next highest integer, which reduces the ability to creep up a point at a time by playing noncompetitive games. If someone who has made 2200 pads their OLM count by playing a large number of noncompetitive games, the count can be adjusted.

1 Like
8

Could you be a bit more specific about what you see as “gaming” the system?

Sure. I mean that, rather than playing in “standard” events to gain the NM title, one could arrange a series of matches or rated events among a small number (that in most cases probably wouldn’t be appropriate to rate in the first place) to beat their friends or colleagues, who either/or (1) have some interest in seeing them get the title, (2) are simply weaker and will lose, allowing the aspiring master to gain points (maybe not one at a time, perhaps a couple against a 1900 for instance, however many points that would be). So, if such small, effectively arranged (one might argue “rigged”) matches are permissible by USCF rules, then the system would effectively be gamed as a loophole would be exploited. Now, is there a rule against this? In the scenario I know of, the issue was brought to the ethics committee and it was thrown out. This is probably because the complaint was all over the place and rife with emotion, but I don’t want to rehash this sort of vindictive approach. I just don’t want to see this sort of thing happen, because it doesn’t seem fair.

9

Thank you, this is very helpful. I like the suggestion that you, seeing the problem, came up with by floating this suggestion:

Some years ago I floated a suggestion that in order to cross a 200 point ratings threshold, a player had to actually compete against players who were one class below that rating or higher. In other words, in order to become a 2000 player one had to have a game in that event against at least an 1800 player. (The Ratings Committee thought the proposal was too complicated.)

And as I mentioned above, I think whatever reform is to occur should be as clear and parsimonious as possible. Hopefully something like performance based norms for the NM as well would be the catch-all to address the entire issue. Or maybe no small matches or something. Finding the right reform that would be widely agreed upon and easy to implement is the trick. But it seems we both agree that something should be done to avoid this sort of unfair play.

10

Thank you for your thoughts on this. I certainly agree that the issue with small towns or remote areas/states in a difficult one. And as I mentioned, I want to avoid titled player “elitism” and, as I work hard to do, make the benefits of the game accessible to all. And as we agree, so long as people don’t gain the highest titles dishonestly and spoil it for the rest of us.

One argument that I mentioned for the small-town conundrum would be to provide resources for the aspiring master to play in tournaments out-of-state. Actually this is where an argument for having the requirement of one strong performance in one open event might make sense. In this way, there would be one big investment (hopefully supported for those who can’t afford the travel and entry fee) to prove that the player is indeed competitive on a national level, as would be expected for someone to attain the “National Master” title. Another possible argument is that small town or remote play against much lower rated players simply shouldn’t confer the title on its own, but that the title of club / town / state champion would be appropriate for this purpose – and that, for the NM title, other arrangements, hopefully as simple as possible, would need to be made.

Come to think of it, if the 2100 something player is so dominant in, say, an Alaska state event, maybe their performance rating would be high enough after all? Not to mention that the event could be open, and therefore the open tournament requirement, if that’s to be the one, would be satisfied. So I think one or the other may work. Does that sound reasonable? Pinning down the specific, most effective and simple reform is the trickiest part, assuming agreement on the need for some sort of basic change is needed.

1 Like
11

It seems reasonable that if the local event was open, not an invitational, then it is publicly open to anyone, allowing in or out of state players to challenge the player trying for NM. But, who will pay over $1000 to fly to Alaska for a tournament? There may be a few, I suppose. But at least local good players may apply.

I fear we in Alaska are going to have rating issues, as most players will have to play within the the 100 (but only 60 active) players in the state, not getting to mix with other states as much to stabilize ratings. But this is a different issue. The same players will end up playing each other in a lot of tournaments, because they are the only players here. I would also wager that a purely Alaska 1700 will not be as good as a Californian 1700, due to the much larger playing pool in CA verses AK.

So perhaps at least one tournament in another state? But would it have to be the tournament where you earned your NM by crossing the magic rating line of 2200? For instance, if you played in the 124th open this year, did well, then went home and finished off your climb to 2200, would that count? Or would you have to end at or above 2200 in an open tournament for it to count? Could it be in any state? Or would it need to be something like the nationals open?

One small argument against the tournament being required to be open, though: Say a senior goes to play in the Irwin invitational, and crosses 2200, since it wasn’t an open tournament, would it not count?

I feel I am rambling now. To be honest, I’d be fine with whatever was decided. Our larger than life state with such a small population may be the exception, rather than the rule that these decisions should be based on.

12

Can you elaborate on this consent to lowering their floor concept. I ask specifically in the context of a floored player who has actually asked to have his floor lowered and was denied, based on the decision-maker’s judgment that he hadn’t played enough games to prove his skills had declined. I personally disagree, but I of course am not an expert and lack authority anyway. Are you telling me that if this guy plays a match (presumably against someone more than a class lower) that he can get his floor lowered? I presume that would still be dependent upon the match result.

13

From the US Chess Rating System (re matches):

  1. Rating floors are not automatically in effect in matches. Instead, if a player has a match result that would lower the rating to below that player’s floor, this will be treated as a request to have that floor lowered by 100 points. If the US Chess office grants this request, the rating will drop below the old floor and the new floor will be 100 points below the old floor.

Note that’s if the person’s ratings would drop below the floor (in practice, someone on a floor who should lose rating points as a result of the match).

Regarding a request to have a floor lowered, it’s a judgment call whether the results show that the floor is really binding (but note that you have to have results). There are people on floors who might play 10 tournaments in a row without the rating ever budging above the floor—to do that, you probably need to be playing at least 100 points (probably quite a bit more) below the floor. Someone with post tournament ratings of 1800, 1806, 1800, 1800, 1812, 1800 is a different story.

14

You do realize that a 2150 playing a 1900 is “betting” 4-1 rating points on the result. The higher rated player would actually lose more on a draw than he would gain on a win.

I’m not sure what the beef is if the games are on the up-and-up. If someone on a floor is playing many games in a small pool, I might be tempted to apply the “match” treatment to the floor since that is certainly going to throw off the rating of the pool compared to the rest of the country. But ordinarily a high rated player is not going to benefit from playing much lower rated players.

2 Likes
15

Here’s how many players have achieved a 2200 rating and NM title by year since 2010.

2010 101
2011 103
2012 133
2013 162
2014 189
2015 151
2016 144
2017 124
2018 130
2019 150
2020 38
2021 67
2022 137
2023 182
2024 61

Looking at these 240 players (I’m not sure why one player is missing), I only find one player for whom the highest player faced in the 90 days before the NM title was earned was a 2000 player.

There are 7 for whom the highest player faced in the 90 days before the NM title was earned was a 2100 player.

There are 19 for whom the highest player faced in 90 days before the NM title was earned was a 2200 player.

All of the rest faced at least one 2300 or higher player, most of them multiple times.

If I expand the search to 180 days, all but 6 faced at least one master, and those 6 all played 3 or more (maximum of 16) experts in that 180 day period.

To me this suggests that there is not much of a problem these days.

3 Likes
16

Wow, the numbers don’t lie, looks like they worked hard for their title. :+1:

Thanks for bringing the stats, Nolan!

17

Looking only at wins, only one had a best win against a player under 1700, another 9 of those NM’s had their highest wins against 2000 players in the 180 days prior to earning the NM title,

Another 18 had their highest wins against 2100 players in the 180 days prior to earning the NM title.

All of the rest defeated at least one master in the 180 days prior to earning the NM title.

Adding in draws, all but 9 had either a win or a draw against a master in the 180 days prior to earning the NM title and all of those had at least one win or draw against an expert.

So I would agree with our friend from Alaska that these new masters appear to be working hard for their titles.

Thinking back to the suggestion I made years ago, I think every NM looked at would have met the criteria I was suggesting.

1 Like
18

So perhaps at least one tournament in another state? But would it have to be the tournament where you earned your NM by crossing the magic rating line of 2200? For instance, if you played in the 124th open this year, did well, then went home and finished off your climb to 2200, would that count? Or would you have to end at or above 2200 in an open tournament for it to count? Could it be in any state? Or would it need to be something like the nationals open?

These are thought-provoking questions. My sense on this has been that so long as an open tournament has been played within a certain range of reaching 2200 then it should be fine. The prospective master would be able to test their mettle in the open event, and if they really weren’t up to snuff, then they’d be pulled back accordingly and would have to climb again. Maybe within the last three events or something like that? Then incremental point gains vs. much lower rated would be precluded since playing vs. stronger players in open events (or at least having a much higher chance of facing them) would continue to pull back a player of non-master strength. These minutiae could be clarified by whichever committee/board would have the last say, so long as they were within their purview.

I was also thinking about possible confusion that may arise when it comes to defining an “open tournament”? (Have to consider all of the candidate moves in reply to fill out and prune the calculation tree! :wink:) Could this simply replicate the problem of one playing for their own org – perhaps it’s still technically an open tournament, if very small? I think one way around this would be to simply rule this out by stipulating that the open tournament can’t be for one’s own affiliate. This would naturally avoid conflict of interest, which is one of the main things to target that I’ve had in mind. Again, of course this has to be kept as simple as possible; I think whatever the final solution is, it should have 2-3 basic points and done.

1 Like
19

Good point @wintdoan. I think the problem is at least two-fold:

  1. Incremental point gains (as you address and perhaps minimize to some extent)
  2. Essentially collusion, as I referred to in my original post. The example that I am aware of, where you don’t just play random lower players, but there’s an “insider” issue happening where the aspiring master plays their friends or colleagues, who may have an interest in seeing the player excel (partially for financial reasons by means of raising the profile of their org and being able to charge more, bring more people in for camps and classes, host simuls, etc.)

So, at least to address the second, I think it may be reasonable to assert some sort of conflict-of-interest clause that avoids this type of internal ramping up of a player’s rating. From what I understand, as I have about “the Borises” (whereupon disciplinary action seems to have been taken), a series of 1 on 1 games vs. a GM were organized. I just don’t think a rationale can be argued for this sort of arrangement other than points being gained for the lower rated player. If they really want to have such as match, it doesn’t need to be rated. I also hear that some people will abuse their TD title by having random games with a friend or colleague and rate it. Frankly, I’ve just heard this from others, so to be fair it would need to be verified. At any rate, the concern is out there and I think it’s fair to address it.

Lastly, I can imagine one thinking, “this is some far-fetched stuff, who is would actually go through the trouble of doing this?” Well, who goes through the trouble of cheating generally? This is just another form of cheating. And, interestingly, the actions that originally brought this issue to my attention involved a player who, lo and behold, admitted to cheating on lichess. So I think it’s all connected.

20

If I expand the search to 180 days, all but 6 faced at least one master, and those 6 all played 3 or more (maximum of 16) experts in that 180 day period.

To me this suggests that there is not much of a problem these days.

Thank you for looking into this data, @nolan. It’s really useful to have something concrete to see just how broad the problem is. And, to be totally fair myself, if the problem isn’t broader than a few anecdotes that I’m aware of, then it would weigh in favor of these few cases being brought before the ethics (or appropriate) committee and addressed this way without the need for reforming the requirements. Again, I’m all for the most efficient solution – one that balances the need for fairness with expediency – avoiding complications and burdens on players, TDs, and those who should (if not already) carefully vet the data and final games of those just reaching 2200 to finally award the title. (I’m not sure if those reviewing the new NM actually look into the final games, but that may be an interesting sort of internal administrative reform, even if it’s just something that’s on the radar but not formalized. Because, naturally, this is the point when one has the greatest incentive for dishonest behavior, when they’re oh-so-close but can’t quite hit that magic number. And again, I know how tough it is at that juncture.)

Beyond the broader data, of course, if the matter is to be put the rest, then probably we’d need to drill into the data a bit. For instance, in the example I’m aware of, most of the behavior that raised flags with me and others was the final run-up to 2200. Hypothetically, perhaps this player is actually of 2200+ strength, but that doesn’t need to be the be-all end-all; they were having trouble actually getting the performance to get there and were probably very frustrated. Many experts are master strength, just as many IMs are GM strength, but ultimately it comes down to OTB performance. (The 2150-2200 climb was incredibly difficult for me, but I fought hard, had a real internal struggle, and finally had performances that I was really happy about to finally break through, so naturally that affects my personal view on achieving the title fairly; not even really a full disclosure, but we all know it’s tough.) So, as this person even admitted to cheating on lichess, I think that it isn’t a stretch to ascertain that they engaged in similarly dishonest behavior to do everything they could within the parameters of their own org – a self-interest issue – to cook up games that would push them through those final 10-15 points. So I really just want to see how common this issue is, and if not, then I’ll respectfully put it to rest.

Given the issue I’ve just mentioned, then, I think it would be interesting to drill into the final, say, 10 games of the players you mentioned above in their final run-up to 2200. Is there a way that I can efficiently take a look at the pages of the players who became masters in, let’s say, 2023 as a relevant recent year sample?