If you’re old enough to remember the Guinness Book of World Records (it still exists, but it’s probably not the mega-bestseller it used to be), you might remember that for many years it claimed that the shortest master game was this “classic” between the French players Gibaud and Lazard: 1.d4 Nf6 2.Nd2 e5 3.dxe5 Ng4 4.h3?? Ne3 0-1.

Terrible chess all around: 2.Nd2 is artificial at best, 2…e5 is an utterly unjustified gambit, and of course 4.h3 is catastrophically bad. But what’s more important is that the game never actually happened. (Which is a pity for anyone who made a bar bet on the game and consulted Guinness for the truth; the whole point of the Guinness Book was to give carefully researched and verified answers to those making inquiries and bets about such matters.) Or rather, it sort of happened - there was a game between those two players, won by Black, with that general tactical idea. But it wasn’t a tournament game, and it went a few moves longer. (See here for a longer discussion and a look at the relevant documentary evidence.)

I was curious, though: maybe Mssrs. Gibaud and Lazard never played the game given above, but did anyone else? Apparently not, at least not according to ChessBase’s reasonably complete Online Database. There are 1655 games that proceeded 1.d4 Nf6 2.Nd2, after which 34 hopeful souls essayed 2…e5, probably hoping to replicate Lazard’s (alleged) triumph. 28 of the 34 white players took on e5, all of the black players played 3…Ng4, and…not one of their prayers were answered! Overall, Black’s results were fine, but not one of the 28 white players repeated (pseudo-) Gibaud’s 4.h3(??).

Now, normally you’d think, “Well, of course not. Why would we expect White to make such a blunder?” That’s a very reasonable thought on your part. The problem, however, is that when you go exploring in the database you’ll find far worse than 4.h3.

For example:

1.f3? e5 2.g4?? Qh4# is there (Pakfolska-Firnhaber, email 2014).

1.f4 e6 2.g4 was played twice(!), but in neither game did Black deliver the mate(!!).

Mercifully, there are no games that started 1.f3 e6 2.g4, 1.f4 e5 2.g4, 1.g4 e5 2.f3/f4 or 1.g4 e6 2.f3/f4.

Not too many cases of “Fool’s Mate”, then, which is encouraging. But “Scholar’s Mate”? There are too many possible move orders, so I decided to look in the Mega Database for games finishing with White playing 4.Qxf7 mate. How many did I find? One hundred and seventy five! Most of the players were unrated, with a significant minority coming from girls’ events. (I’ve certainly seen boys fall for it, but apparently their coaches did a better job of getting it out of their system before the played in database-worthy events.) But it was not only very young newbies who fell prey. Quite a few had respectable club ratings in the 1700-1800 range, and there was even a player rated 2334(??!) who (allegedly?) allowed it. Turning the tables, another 51 players on the white side allowed 4…Qxf2#. So it becomes increasingly surprising that no one has replicated Gibaud-Lazard.

Another elementary trap (and back to the Online Database as our source): 1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.Bf4 Nc6 5.Nf3 Bb4+ 6.Nbd2 Qe7 7.a3 Ngxe5 8.axb4?? Nd3#. This comes in three versions. One we just gave, a second sees Black play 7…Ncxe5 instead and give mate on the next move, and the third has White play 8.Nxe5 (it could be the knight from c6 or the knight from g4 - it comes to the same thing) Nxe5 and then walk into 9.axb4?? Nd3#. Here are the numbers for this one: Version 1: 94 cases. Version 2: 15 cases. Version 3: 11 cases. To Black’s credit, the mate was executed in every instance.

One more: 1.e4 c6 2.Nc3 d5 3.Nf3 dxe4 4.Nxe4 and now either 4…Nf6 5.Qe2 Nbd7?? or 4…Nd7 5.Qe2 Ngf6?? shows up 19 times (most famously in the 1950 game Keres-Arlamowski), and all 19 times White played 6.Nd6#.

I’m sure the examples can be multiplied; if any especially egregious cases have caught your attention, feel free to mention them in the comments.