Free online Chess server. Play Chess now in a clean interface. No registration, no ads, no plugin required. Play Chess with the computer, friends or random opponents.
Create a game Tournament Simultaneous exhibitions
Chess basics Puzzles Practice Coordinates Study Coaches
Lichess TV Current games Broadcasts (beta) Video library
Players Teams Forum Questions & Answers
Analysis board Board editor Import game Advanced search
Sign in
  1. Forum
  2. Game analysis
  3. I need help with my Philidor study

Hi all,

I've recently done some studying on the Philidor defence which I have poor score against. I might publish my extensive study later, but for now I want to discuss just a part of it. After 1.e4 1.e5 2.Nf3 d6 3.d4 Nd7 4.Bc4 black should respond with 4...c6. However, it's easy to play 4...Nf6? which according to my analysis gives the white two great options. (As a side note, 4...Bd7? is even worse and 4...h6? allows white to launch a brilliant attack including a bishop sacrifice on f7.)

White can play the safer 5.exd5 which wins a clear pawn. Black recaptures with the knight: 5...Nxe5. (Not with the pawn because of 5...dxe5?? 6.Ng5 and the f7 pawn cannot be defended.) Game continues 6. Nxe5 dxe5 7.Bxf7+! Kxf7 8.Qxd8 Bb4+! 9.Qd2 Bxd2 10.Nxd2 and white has clear pawn advantage.

The other, more exciting option is 5.Ng5!? Then 5...d5 is forced and we play 6.exd5. 6...Bb4+ and 6...Bd6 give the white a 0.4 advantage according to Stockfish, but let's suppose black chooses the third alternative 6...h6 which is +0.5 according to Stockfish. We could retreat with 7.Nf6 and have that slight lead, but more interesting is 7.Nxf7! The game continues 7...Kxf7 8.d6+ Kg6! (8...Ke8 9.Qd3 9...e4 10.Qg3 11.g5 dxc7 Qe7= white has three pawns for a knight) 9.Qd3 e4 10.Qg3+ Kh7 11.Bf7?! g5 dxc7 Qe7. White still has some ideas, but they are becoming more desperate.

So you can choose a safe 5.exd5 whcih gives white a clear edge or the exciting 5.Ng5!? My question is, why is EVERYONE playing the rather passive 5.Nc3? A great majority of the games in the tablebase continue 5...Be7 6.O-O O-O. The white is better, but is he better than after 10.Nxd2 in the 5.dxe5 line? I don't think so. The tablebase also tells me that by playing 5.Nc3 you lose 24 percent of games, draw 38 percent and win 38 percent of them. I'm sure that on top level white would do better than that if he is a pawn up in a peaceful position, right?

There is clearly something I don't understand. Help appreciated.

It is the wrong move-order either way. Hanhams Nd7?! is full of dubious lines. Every good player starts: d6, Nf6, e5, Nbd7, Be7.

So you mean
Philidor is not popular at top level.
Those who play Philidor play 4...c6.
So the position only occurs at lower levels where the white players are not very skilled either, or just lazy and happy to transpose to a position they know with 5 Nc3.

Well, aside from the relative merits of the Hanham, there are a couple points.

First, 4...Nf6? might be "easy" to play in some sense, but it's not a common response even in the DB of lichess games with no rating or time control filters, and is basically non-existent in OTB master play.

Assuming black plays that bad move, then yes, 5.dxe5 is the strongest option.

However, I'm not sure why you say "EVERYONE" plays 5.Nc3.

4...Nf6? just isn't played at master and up OTB (exactly 1 game in the lichess master DB, for example), so it's a bit silly to talk about what "everyone" plays after that.

Even in the unfiltered DB of lichess games, where weak moves like 4...Nf6? are to be expected more often so that we actually have a distribution of responses, dxe5 is the most popular response to 4...Nf6 by a decent margin.

My best guess is that you used another opening explorer for this, one that counts transpositions to known positions in the count of games for each move (in other words, they count the number of times the position AFTER the move has arisen, not the number of times that move was played in THIS position).

So, for example, after 4...Nf6?, 5.Nc3 transposes to a much more common line, so if you used an opening explorer that counts such transpositions (which is most of them), then it'll look like 5.Nc3 is common, when it actually isn't.

It just transposes to a more common line (1.e4 e5 2.Nf3 d6 3.d4 Nf6 4.Nc3 Nbd7 5.Bc4), in which none of those tactical lines ever actually worked for white because of the move order.

You can see this quite clearly if you use, e.g. chesstempo, which has the transposition "feature". There's only one game in the 2200 vs 2200 with 4...Nf6?, but if you click on it, all of a sudden there are thousands of games in which indeed, nearly "everyone" plays Nc3.

That's just the transposition effect (obviously you can't have thousands of games played from a position that's only been reached once); sometimes it's nice to know transpositions to a common position are available, but sometimes, like here, it's just really misleading.

It's why I've always preferred 365chess's explorer, and more recently lichess', since they handle transpositions the way I prefer.

Hopefully this helps clear up the confusion :)

Thank you @a_pleasant_illusion. That's it. I mislooked AND misinterpreted some databases. This definitely cleared the confusion and I can happily play 5.dxe5 if the position ever arouses. At my level it just might so you need be aware. :) They usually have learned that black plays 4...exd4, though. Then it's hard to attack as black doesn't have any clear weaknesses. I've looked into main lines 5.Nxd4 and 5.Qxd4 but also looked if gambits like 5.Bc4 or 5.c3 would work.

6...Nb6? won't quite work like that @tpr. White is happy to play 7.dxe5. Then 7...Nxc4 8.exf6. The white f6 pawn is a beast and there is an open e-file at black king who is unable to castle. Or if 7...Nxd5 8.Qf3 threatens mate on f7 and prevents 8...Nxc4. Black is a pawn down and white has a fierce attack. There are even other ways to refute 6...Nb6?

This video might help.