Openings like this really test your chess skills. It sharpens your senses, you concentrate on good lines, because your opponent has already begun conducting an attack! A real hustler's opening.

[Event "?"]
[Site "Paris"]
[Date "1980.??.??"]
[Round "?"]
[White "Gedult, David"]
[Black "Przibyczevski, P"]
[Result "1-0"]

It would not be unusual to encounter duplicate games when a person plays a particular line frequently. For example, I play QGA a lot, and so have encountered the following position numerous times:

1. d4 d5 2. c4 dxc4 3. e3 b5 4. a4 c6 5. axb5 cxb5 6. Qf3 resigns

since 6. ..Nc6 (best move) 7. Qxc6+ Bd7 8. Qa6 and white is up a piece and the win is a matter of technique.

<whiteshark: I once read all of his games were rapid games with 15 mins. So, really worth studying.>

There are 28 games in the database, so it is obvious these are not all his games. As for the 100% win record, this is not a competition performance evaluation. I could present several hundred wins with some very interesting mates and so forth, but that alone is not an accurate measure of my playing strength. Strength is only measured against strong competition, but the games in the database for the most part seem to be against middle class players, not experts and masters. And if only one's wins are provided, it would not be a realistic representation.

Deimer considered him a "talented student".
Yet there are no games between Deimer and Gedult in the database. He played correspondence with Gunderam, yet none of those games are in the database. He embraced the opening repetoire of Gunderam, Deimer, Meises. He liked to write about chess as much as he liked to play.

What can we conclude from this?
He no doubt spent a lot of time analyzing the BDG and KGA/KGD as well as many other gambit lines that were popular in the Romantic era of chess. Such being the case, one would expect him to have a good collection of wins with these lines. Yet his play is not merely artificial. He demonstrates at least an expert/candidate master level of play.

Therefore, i would have to guess that what we have in the chessgames database is a catalog of opening mistakes. That is to say, his personal games that illustrate the opening errors in the lines he preferred, and probably about which he wrote chess articles. Since they were published, they were recorded, and therefore in the database. And if he was not an all around "master", he seems to be at least a class A (1900-1999) or candidate master (2000-2199)

The moment I began having my doubts was after seeing a whole series of Gedult games starting out 1.e4 e5 2.Nf3 f5 3.Bc4 fxe4 4.Nxe5 Qg5 5.d4 Qxg2 6.Qh5+ g6 7.Bf7+ Kd8 8.Bxg6! and many turbulent sacrificial ideas. Only a few players were aware of this kind of generally unknown line, and Gedult suddenly played against half a dozen Latvian gambit experts in one playing session ! Next time, a whole series of Kings Indian Four pawn attacks with a early e5.. popped up from nowhere... Paris players told me that Gedult often suggested to analyse some interesting line by playing a range of blitzgames..

So these games really prove that you should not believe everything that you see ! It is not clear what has been played (in blitz) and what has been made up. But keeping this in mind we can also be grateful to a certain extent that Mr. Gedult has composed these chess fairytales for us !