The Urusov has been popular among attacking players for nearly 150 years. Adopted by Schlechter, Tartakower, Caro, and Mieses, the opening claimed victims among the best defenders of the late 19th and early 20th centuries, including Steinitz and Lasker. By 1924 there was enough interest in the line that a thematic tournament was organized in New York featuring Marshall, Torre, Santasiere and four local New York masters. After the Dimock Theme Tournament the Urusov’s popularity waned until correspondence players began exploring the opening's many forcing lines and Yakov Estrin (World Correspondence Champion from 1975 to 1980) published several monographs that carried the analysis well into the middlegame. Estrin's analysis revealed an equalizing method for Black (with Panov's 4....d5) and suggested that some of the lines might end in equality with best play.
Recently I decided to give it a try in an LSS tournament because I had looked at the games and analysis and became intrigued by its possibilities. Also, I wanted to experiment with the Rybka engine’s Monte Carlo Analysis Method, something I had not been able to do previously because I didn’t have a Rybka engine. I succeeded in playing the Urusov in two games, winning one and the other is still in progress. I am fighting for a draw in that one. This post contains only some of the analysis that resulted.
In one game after 1.e4 e5 2.Bc4 Nf6 3.d4 exd4 4.Nf3 Black chose 4…Nc6 which according to the engine opening book is the most popular, but I think it’s already mistake. White’s 4th move threatens to advance with e5, forcing the Knight to e4, where it is vulnerable to attack. So while 4…Nc6 is the most popular, it doesn't seem logical in view of the fact that it does not prevent White from playing the move he wants to play, namely 5.e5. Monte Carlo Analysis statistics seem to bear this out as White scored +180 -104 =92.
In the other game Black chose the Panov line: 1.e4 e5 2.Bc4 Nf6 3.d4 exd4 4.Nf3 d5! 5.exd5 Bb4+. On move 8, after a lot of deep positional analysis and Monte Carlo analysis, I selected the less popular 8.Nc3 instead of 8.bc3. MC analysis showed 8.Nc3 as yielding better results for White, but the further course of the game would seem to indicate that two factors influenced the outcome: the shallow depth (7 ply) and the fact that Deep Rybka 4w32 is not as strong as Houdini and Stockfish engine, resulting is less accurate results.
Against Panov’s 5…Bb4+ I was unable to avoid a slightly inferior position. In fact, as late as move 20 or so I did a Shootout with Houdini 2 and White scored +0 -1 =3. Hardly conclusive, but it illustrate the difficulty White has against Panov’ idea. I won the game where my opponent played 4…Nc6 and am struggling to hold the draw against Panov’s 5…Bb4+.
Conclusion: I think the Urusov is good to play over the board for those that like gambits. You could find yourself in trouble IF your opponent is booked up on the Panov idea, but that’s probably not going to happen. It could also be played in correspondence chess on places like ICCF and LSS, but with engines being allowed on those sites, you will have to search really deep to find a way for White to equalize if you run into the Panov variation.
If you are looking for an opening project to tinker around with you can try searching for something for White after 1.e4 e5 2.Bc4 Nf6 3.d4 exd4 4.Nf3 d5 5.exd5 Bb4+! Who knows, you could get a variation named after yourself.