The Format for the Chess Olympiad is Stupid

I don’t know why the chess community can’t get the following extremely simple concept through their head:

“Major chess events should conclude with a climactic showdown between the two top players or teams”

This extremely basic concept, which is used in nearly 90-95% of all major sporting competitions, is seemingly lost on the chess community. In fact to even suggest that “hey, maybe instead of Team 1 playing Team 4, and Team 2 playing Team 7 in the final match, how about they just play each other?”, you’ll be treated to a litany of people spamming your blog post to inform you that you’re the biggest idiot in the world.

Will these people write to MLB, NFL, NBA, NHL, the Olympic Games, The World Cup, The Champions Cup, The Euro Cup, The NCAA and basically every major sporting competition except for National Soccer Leagues, to tell them how dumb they are?

We just witnessed the Olympiad ending in ridiculous fashion, with USA playing Canada, while Ukraine played Slovenia. Neither team had any chance of getting first place, yet here they both are, determining which team will be Champions of the world. But that was not the most ridiculous part of the final round. The most ridiculous part was after the USA and Ukraine unsurprisingly dispatched their much lower ranked opponents, the result then came down to the match between Estonia and Germany.

Germany and Estonia came into the final round of the event in 54th and 70th place. All eyes were on them as because of some absurd tiebreak format, the USA needed Germany to win in order to win the gold medal.

Let me counter the normal argument that the chess fans all scream in unison:

“But wait Greg, in the NFL there are dumb tiebreaks all the time that determine what teams make the playoffs, so how is this not the exact same thing”

It’s not the exact same thing because we aren’t talking about which borderline teams just squeak into the postseason. We are talking about which team will win the gold medal! You don’t determine who wins the Super Bowl by watching the Cleveland Browns play the Tennessee Titans.

Here is exactly how the Olympiad should work:

There should be 9-11 rounds of play. At the end of these rounds, the top four teams go into a four team playoff, with 1 vs 4 and 2 vs 3. Do I care that the fifth place team didn’t quite make the cut because of a tiebreak? Of course not! I’d rather see a tiebreak affect 5th place than 1st place! Let the teams fighting for first battle it out head to head.

If the playoff matches are drawn, there is a team tiebreak match at rapid time control. If it’s drawn again, speed up the time control again. Eventually someone will win. Then the two winners play in a grand championship match.

What I have just proposed above is exactly how any sane sporting organization would run an event like this. But it’s not the way chess does it, and they haven’t for years, so people will complain about how the current system is better. The current amazing system in which the final round consists of a bunch of mismatches and then sitting around and watching the 70th place team to see who’s the champion, if we can even figure out the tiebreak system to begin with.

Can the chess world for once just admit that the rest of the world isn’t insane or delusional, and actually knows how to run sporting events? Ukraine and USA should have played today. They both went through the tournament with a 10-1 record! And the winner of their head to head match should have been champion. This is absolutely basic.

But at least for all the U.S.A fans, the tiebreaks fell on our side this time.


12 thoughts on “The Format for the Chess Olympiad is Stupid

  1. The fact that blitz can determine the outcome disqualifies this scheme in my mind. (I feel the same about penalty kicks in soccer, and all other chess events that use rapid/blitz/armageddon to determine the winner.) The SB tiebreaks are a mathematically fair and legitimate way to determine who wins a large swiss event.

    Why is it “fairer” to have two tired teams (maybe) have to play blitz games to determine a winner? That’s a far worse way to decide who’s “the best” in my opinion.

    Your scheme is fun for casual spectators, but bad for actually measuring what team should win the gold.

    Let’s let chess be chess instead of trying to make it into American football.


    Liked by 1 person

    • We can expect that a playoff (even one including rapid/blitz) will be far more correlated with the skill of the teams than an arbitrary formula for tiebreaks.

      Tiebreak systems in these kind of tournaments tend to come down to whether the bad team you were paired with early in the tournament does better than the bad team your opponent played against, which is pretty much just a coin flip.


      • The formula is *not* arbitrary and measures exactly what you say, i.e. who played teams that had a better performance *in that tournament*. Sure, there’s a bit of randomness to it, but much less so than (potentially) four blitz games.

        Especially if there’s a four team playoff, as Greg proposes, the chances that the 1st place team loses to the one of the five(!) teams tied for 4th place in the first round are too high, and that would suck.

        I don’t think the best team should be at a higher risk of losing just so it can be more entertaining, and we have a quicker final result. The best team should win the tournament.


  2. I have to respectfully disagree. The Olympiad is exciting precisely because you have underdog teams competing with the big favourites. I’d be more on the edge of my seat to watch a huge underdog pull off a stunning victory to deny one team gold than I would be seeing a rematch of US vs Ukraine. One reason why some super-tournaments get stale is that the same players play each other again and again. I get that some people don’t mind seeing Cleveland and Golden State duke it out every NBA finals. Not for me…

    Moreover, your comparison to other sports is flawed. There is no individual competition elements to the sports leagues you mention that is in any way analogous to chess. Secondly, those leagues is choose the best teams, but with good competition among the teams involved.Your proposal about making the winner decided by a playoff devalues the early rounds to an even greater extent than any other sport. Even in the NFL with 16 regular season games you play ~40% of the other teams during the regular season. The closest thing we have to your proposal might be the recent change of NCAA football to add to the BCS system with a playoff. Did that turn out better, perhaps. Did it drastically improve the “stupid” system for you? For me it didn’t matter that much.

    Lastly, I would argue the point of the Chess Olympiad reaches more beyond the goal of choosing a winner compared to the other sports league you mentioned. The team aspect of the Olympiad is of course about the drama and intrigue of what country does well and who might win. The Olympiad is also is about getting the chess world together (not just potential winners). Its about seeing the diversity in chess – the young up and coming talents, the old guard reigniting the spark they showed decades ago, the huge upsets, and so much more.


  3. Great article, Greg – and immediately, we see the arguments between those who want a champion who rises to the challenge in the crucial moment vs. those who want a champion who is statistically superior over the length of an entire season (and which, in the case of this year’s chess olympiad, won the head to head USA-Ukraine match, too – though it still came down to Germany vs. Estonia for that to have held up).

    It has always struck me that the sports which culminate in a mano-a-mano (whether individual or team) final for the title are seeking whoever can come through at that last critical moment, after reducing both sides to exhaustion – nothing left but will and instinct. The lengthy seasons of most professional sports are what reduces the players to that level, so we can find out who has the pedigree when the moment comes. Compare, for instance, the Chicago Bulls of Michael Jordan, with the Golden State Warriors who eclipsed the Bulls’ 72-10 NBA record regular season record last season- but who fell in the championship series and went home without the big prize. Or a boxer who gets knocked out in the 15th round while ahead on points in a world heavyweight title bout. Instead, most sports want to see who scores the OT winner in game 7 of the Stanley Cup final, or who comes through in the 9th inning of the last game of the World Series (and on the other side, who lets a ground ball go between their feet, to lose it all).


  4. There are two issues that muddy this debate when they’re not clearly set out: (1) which format is the most exciting for the spectator; and (2) which format produces the most just winner. I don’t think any one can seriously debate the first point — a system that matches the top two teams in a winner take all match — and follows it with tiebreakers at ever-decreasing time control is the most exciting. With “most exciting” defined as most exciting to the most people. [There will always be outliers who try to debate the Premier League or the Bundesliga format is more exciting, but they are a clear minority.] It is also true, however, that it is *because* it is the most exciting, that the other sports leagues have adopted that format. More excitement equals more fan engagement which equals more revenue. The argument can be made, that this is reason enough for chess to adopt that format. More revenue results in more professionals, which results in greater popularity, which grows the game, and so on.

    It is also true, however, that simply because it is the most exciting, doesn’t mean it is the most just. In any format where the teams are not competing against the same opponents, arguments can easily be made that the top 4 or top 2 rose to the top, in part, because of lucky draws. You can only eliminate that argument by ensuring all teams play the same opponents, [Which in the case of the Olympiad, is impossible.] But it’s at least a fair argument that the current system is as just (or more just) as the system you propose. In either scenario you are making determinations of the better team(s) based on something other than head-to-head competition.

    Personally, I prefer your system a whole lot better, mostly because it is more exciting, but also because it is as relatively just, IMHO, as the current system. I just think this debate could use some framing to avoid the arguments going past each other.


  5. Everyone who disagrees with Greg is dumb and doesn’t want chess to grow in popularity. I don’t understand how anyone could support the current format over a playoff. How is a blitz playoff less worthy of declaring a winner than a game between Germany and Estonia. Blitz is still chess, no matter how much people say it isn’t because they suck at it.


  6. Well, I only partially agree.
    Yes – the current system sucks; but no – playoffs is not the solution.
    The main problem today is the stu**d SB tiebreak. I TOTALLY agree with Greg that this is an impossible method, and that the negligible GER-EST match is the last thing that should decide the tournament winner.
    But – playoffs is not the solution, due to many reasons that were stated here above (unfair, a “black” day can ruin everything, relies on luck in many occasions etc.).
    The way to go is just change the tiebreak system. The 2nd tiebreak, which is the total number of points won by the team (maximum of 4 in each round) should be the first tiebreak. This actually determines how well you played, and benefits a 4:0 win over a narrow 2.5:1.5 victory. In many aspects it is similar to goal-difference in football (not american), where the team that scored more and conceived less has an advantage when the overall score is tied. In fact, until a few years ago this was the actual score that determined the world championship – they counted the accumulated points throughout the tournament.

    This is a good method since (a) it eliminates any effects of last-round match-ups who don’t involve the actual title (or medal) competing teams; (b) it is simple to calculate and each side knows before the last round what are his options; (c) and it encourages attractive and offense-oriented chess because any point can help in future tiebreaks (which means that even if my teammates took 2.5 points in three matches, I will still play to win in order to increase our “goal-difference”).

    If you ask me, even the 2nd tiebreak should not be the SB, but the head-to-head result – but this is less acute since it will be rather rare to use it.


    • I’m sure there are reasons why the team points was not made the second tie-break. It might even be a matter of initial round matchups. But each system has their pros and cons. I’m also in favor of using the team points the second tie-break.


  7. No need to have “blitz” tiebreaks in a match.

    Just give one side all Blacks, but draw odds. Not sure whether it is 4 or 5 player teams that this becomes essentially 50-50, but it’s way better than the blitz solution.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s