Jerome Gambit: Antonio-bot (Part 1)

After matching wits with Chess.com bots Deadlost and Grandpa Gambit, I decided to challenge the "advanced" bot, Antonio.

Antonio prides himself on being a balanced player. He doesn't win every game, but don't expect to beat him without a fight.

Antonio did very well against the Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) - up until an ill-advised return of sacrificed material. After that, trouble snowballed.

Comments by Antonio are in brown.

perrypawnpusher - Antonio-bot

Chess.com, 2023

1.e4

Sicilian today, or something else?

1...e5

Let's play an open game.

2.Nf3

What are you up to?

2...Nc6

"I don't believe in psychology. I believe in good moves." - Bobby Fischer

3.Bc4 Bc5 4.Bxf7+

I believe in psychology and good moves - Rick

Take that one off the board.

4...Kxf7 5.Nxe5+

Is that my king you're pointing at?

5... Nxe5 6.Qh5+ Ke6

7.Qf5+ Kd6 8.f4 Ng6 

According to The Database, this is the reply to 8.f4 is seen about 1/6 of the time. Black saves the Knight. 

9.Qd5+ Ke7 10.Qxc5+ d6 

Do you see my plan?

11.Qe3 

This is the 10th time that I have arrived at this position.

I suspect that after each of those games, the computer suggested that 11.Qc3 was better. I have yet to take that advice.

11...Kf7 

Antonio moves his King out of the eventual line of fire, the e-file.

12.O-O Nf6 13.f5 Ne7 14.d4 

Here we have the classic Jerome Gambit battle of White's 2 extra "Jerome pawns" against Black's extra piece.

Stockfish 15.1 (34ply) rightly assesses the defender as being about a little bit less than a pawn better.

14...Ng4 

Suddenly the bot shows human tendencies: Let's attack the Queen!

Consistent would have been 14...Re8 with still the better game.

15.Qg5 

Stockfish 16 prefers 15.Qb3+ here, but I am never sure if this is a strong move, or just a distracted, harrasing one. 

15...Nf6 

No harm, no foul? The computer thinks so.

16.e5 

There is a tactical "hole" in this idea that neither I nor my opponent noticed. Black now has 16...dxe5 17.dxe5 Qd4+!? 18.Qe3 Qxe3+ 19.Bxe3 Nxf5 20.Bd2 Ng4 21.h3 Nxe5 22.g4 Kg6 23.gxf5+ Bxf5 and would remain ahead. 

16...Ne4 

Let's attack the Queen!

17.e6+ 

This move is okay, and has attacking ideas behind it, but more accurate would have been 17.Qh5+ Kg8 18.f6 breaking lines open.

17...Kg8 18.Qh4 Nf6 19.Bg5 

This is a comfortable position for White, if only because it is an uncomfortable position for Black.

[to be continued]

Comments?

The question has come up, Why doesn't this blog accept Reader comments?

The answer, quite simply, is that it does.

Over the years I have received 340 comments, 329 of which have been posted - the remainder being spam (it's everywhere), and thus not posted.

It is also likely that some additional comments have been rude, crude or socially unacceptable, and thus immediately remanded to the trash can icon; but those have been few and far between.

Readers can scroll to the bottom of any post, find the small No comments phrase (indicating that there have been no comments on that post, yet) and click on it. They can then follow the directions and comment in the text box.

It is also possible to find my email address (richardfkennedy@hotmail.com) in the About Me sidebar and contact me that way. Many readers have.

As I have noted there,

I am always interested in receiving games and analysis: as old as 1874, when the opening first was published, or as recent as today -- casual or serious, blitz or classical time settings, human or computer (or both).

Jerome Gambit: Grandpa Gambit (Part 3)

                                                

                                 [continued from the previous post]

perrypawnpusher - GrandpaGambitBot

Chess.com, 2023

Neither player in this human vs bot game seems interested in active play. Stereotypical.

28...Re6 

Stockfish 16 prefers the more active 28...Re4

29.Re2 

Now was the time for the plan 29.h3 Kg6 30.g4 Kf7, so Black could have tried 29...Rg8 instead.

29...Rxe2 30.Kxe2 Re8+ 31.Kf3 h4

32.g4+ Kg6 33.h3 Re7 34.Rf2

There was probably a smarter way for me to play this position, but this move suggests that my plan was "duck and cover".

34...f5

This gives away a pawn and more. Not that I noticed.

35.g5 

At the time, I was convinced that this was a way forward.

Readers skilled in Rook-and-pawn endgames will understand Stockfish 16's post-game analysis; 35.gxf5+ Kxf5 36.Rg2 Re1 37.Rg5+ Kf6 38.Rxd5 Ra1 39.a3 Ke6 40.Re5+ Kf6 41.Kg4 Rg1+ 42.Kxh4 Rf1 43.Kg4 Rg1+ 44.Kf3 Rh1 45.Ke4 Rxh3 46.a4 Rh2 47.Rb5 Rc2 48.Kd5 Ke7 49.a5 bxa5 50.Rxa5 Rxb2 and White is much better. 

35...Re1

The active Rook (as well as my nerves) guarantees the draw.

36.Re2 Rf1+ 37.Rf2 Rg1 38.Rg2 Rf1+ 39.Ke3 Rh1 40.Kf2 

Even game.

It is a bit humorous to note that Black will now gain nothing by capturing the h-pawn, as the Rook will be interned behind enemy lines, e.g. 40...Rxh3 41.a4 Rh1 (41...Rg3? 42.Rxg3 hxg3+ 43.Kxg3 White is winning; 41...Rd3 42.Ke2 Rh3) 42.Rg1 Rh3 43.Rg2 Rd3 44.Ke2 Rh3 etc. ensuring the draw.

40...a5 41.Rg1 Rh2+ 42.Rg2 Rh1 43.Rg1 Rh2+ 44.Rg2 Rh1 

Draw

Well, things fizzled out for me, and this time I was not able to take advantage of my advantages in the endgame.

We need a rematch.

Jerome Gambit: Grandpa Gambit (Part 2)

 

                                 [continued from the previous post]

perrypawnpusher - GrandpaGambitBot

Chess.com, 2023

Continuing the grandpa (that's me) vs Grandpa (a Chess.com bot) chess game. 

13.Qe3+

Much more interesting was 13.Qg5+ (strongest) Ke8 14.O-O!?, which I hadn't even considered. Black should probably then exchange Queens (blunting White's Kingside pawn majority), and after 14...Qxg5 15.fxg5 his position wouldn't look that bad - although the computer still assesses the position as 4 pawns better for White.

13...Kf8 14.Nc3 Qe8 15.d4 Nf6 16.O-O Qxe3+ 17. Bxe3 Kf7 

Black's pawns are still a mess, although White's d- and f-pawns are doing his "bad" Bishop no favors, either.

What should my plan have been?

Certainly d4-d5 comes to mind (now, not then) to give the Bishop more breathing room. Occupying the e-file, while drawish (Rooks can be exchanged) is another possibility. Trading off White's "bad" Bishop for the "good" (better) Knight as also possible.

Sadly, what follows, instead, is a mish mash of partial ideas.

18.Nb5 d5 19.Rae1 b6 20.Nd6+ Ke6 21.Nxc8 

Trading off the "good" Knight for a "bad" Bishop because I can meet the next move with c2-c3, further blocking in my own Bishop...

21...Raxc8 22.c3 Ke7 23.Bf2+ Kf7 24.Bh4 Rhe8 

Perhaps Black calculated a draw after Rooks were off the board.

In any event, it was time to give up the "minor exchange" - trade my "bad" Bishop for his "good" Knight.

25.Bxf6 gxf6 26.g3 

Protecting my f-pawn and limiting his h-pawn. After the game, Stockfish 16 had a mind-numbing suggestion that indicated a balanced game: 26.Rxe8 Rxe8 27.Kf2 f5 28.a4 Kf6 29.g3 Re7 30.Kf3 Re4 31.Rh1 Re8 32.Ra1 Rh8 33.Kg2 Re8 34.Kf1 Rb8 35.Kf2 Re8 36.Re1 Re4 37.Kf1 Kg6 38.Ra1 Re8 39.Rd1 Kf6 40.Re1 Re4 41.Rxe4 dxe4

26...Kg6 27.Kf2 Kf5 28.Kf3 

Black's King cannot invade the Kingside. A few ideas for him: put his Rooks on g8 and h8 and advance his h-pawn; or put the Rooks on b8 and c8 and pursue a "minority attack" with his pawns; or park a Rook on e6 or e4 and double Rooks on the e-file - but which square he chooses makes a difference.

White should think about h2-h3 and g3-g4. 

                                                           [to be continued]

Jerome Gambit: Grandpa Gambit (Part 1)

 

After defeating Deadlost (see "Jerome Gambit: Deadlost [Parts 1, 2, and 3]), a chess bot at Chess.com, I decided to challenge the Grandpa Gambit bot.

Grandpa Gambit is a lifelong chess player known to regale his opponents with a chess anecdote of his past…or two. Ready to take a trip down memory lane?

I played the Jerome Gambit, of course, and this may have offended my opponent (if such a thing were possible), as he was silent throughout the game, posting not a single remark.

perrypawnpusher - GrandpaGambitBot

Chess.com, 2023

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 

7.Qf5+ Kd6 8.f4 

A quick look at The Database shows that I have done well against this defensive lineup, scoring 79 - 14 - 1, that is 85%.

8...Nc4

This is an odd move. It is hard to find the idea behind it, other than escaping the attack of White's f-pawn.

If the Knight were to retreat, it should have gone to c6, i.e. 8...Nc6.

Better was a Queen move, either 8...Qh4+ 9.g3 Qf6 or 8...Qf6 directly.

Possibly the best move was 8...Kc6, simply abandoning the Knight, followed by 9.fxe5 Nh6 or 9.Qxe5 Nf6. 

Perhaps the issue was that the dangers ahead lay beyond the computer's depth of search.

9.Qd5+ Ke7 10.Qxc5+ 

At this point, Black resigned in perrypawnpusher - KanatolyAsparov, 4 16 blitz, FICS, 2011.

10...Nd6 11.e5 

White will recover his second sacrificed piece, with advantage.

11...h5

Another unusual move, reminiscent of the choice of the early dedicated chess computer game, Chess Challenger 7. As I mentioned in "Jerome Gambit: The Science of the Draw?! (Part 1)"

when it assessed its King to be safe (wherever it was) and the position balanced, couldn't "think" of what to do, and so would advance its Rook pawns...  

About 15 years ago I challenged the CC7 with the Jerome Gambit, and won in 20 moves.

12.exd6+ cxd6 

Here I was a pawn ahead, but Stockfish 15.1 (36 ply) sees White's advantage as four times that. 

The danger for me, as always, was inattention and complacency.

[to be continued]

Jerome Gambit And Professor Z

It is quite possible to look at the Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) for the first time and tell yourself "This is so far out, it must be fiction!"  

About a year ago, thanks to information provided by Yury V. Bukayev, we actually looked at the intersection of the Jerome Gambit and fiction, in considering Tommaso Castellani's Italian language novel, Il professor Z e il segreto del triangolo, Un giallo matematico, (Professor Z and the secret of the triangle, a mathmatical mystery), published by Edizioni Dedalo, 2022  .

In the book, young Ivano shares his interest in the JeromeGambit  with a friend. See The Jerome Gambit and the Secret of the Triangle and Jerome Gambit, Triangle, Infinity, Hypercube. 

As an exciting large success, out of more than 300 books considered in the Premio Nazionale di Divulgazione Scientifica 2022 competition of science books for younger people, 50 titles made it to the preliminary stage, and Tommaso Castellani's Professor Z and the secret of the triangle was one of them.

Congratulazioni!

Jerome Gambit: Deadlost (Part 3)

[continued from the previous post]

perrypawnpusher - DeadlostBot

Chess.com, 2023

We continue my Jerome Gambit game (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) against Chess.com's Deadlost.

Deadlost's comments below are in red.

23.dxe5 Nd7 24.Bf4 

Is it just me or has the detached, ironic humor of the 2010s run its course? And metatextual references have lost all potency as writers increasingly recycle references to elicit immediate, and unthinking, reactions in their audiences? The Hollywoodification of art has homogenized our unique sensibilities to produce a bland, repeatable template designed to appeal to the widest audience possible. Treating irony as a universal experience, writers now present their stories as something to be mocked. And as such, audiences have developed

calloused, nihilistic perspectives which may never again accept earnest or thoughtful storytelling. Anyway... chimichanga! 

24...Rae8 25.Re1 Rhf8 26.Nc2 

Instead, 26.Nb5 might have been a bit better.

26...Bxa2 27.Nd4 Ndxe5 28.Nxc6 Nxc6 29.Qxe8 Rxe8 30.Rxe8

I think I have mentioned that despite the fact that I play sacrificial attacking openings like the Jerome Gambit, I am often drawn to exchanging down to a much simpler, winning, endgame. 

30...Bf7 31.Re1 h5 32.Bxc7 

Up the exchange and a pawn. That should be enough.

32...Bc4 33.Kf2 a6 34.Kg3 Bb3 35.Be5+ Kg6 36.Kh4 Kh6 37.Bf4+ Kg6 

38.Re8 Bd5 39.g3 Bh1 40.g4

Finally! An idea on my part.

40...Bf3 41.gxh5+ Bxh5 42.Rg8+ Kf5 43.Kxh5 Kxf4

Sigh. Play it out.

44.h4 

Yury V. Bukayev points out that 44.Rg7 would be simpler, as both of Black's pawns are vulnerable and will be captured. The Queenside action would be a good demonstration of the advantage of the exchange in the endgame, as the Rook would out-perform the Knight.

44...a5 45.Kg6 

This move blunders away the exchange, although the game is still won for me. I was getting sloppy, figuring that I had already marked up "1 - 0" on the scoreboard. Deadlost did not overlook.

45...Ne7+

I think that's called check. Let me get out my chess dictionary... yup. That's check.

46.Kf7 Nxg8 47.Kxg8

Congrats! You are the 1 millionth person to capture my knight this month. Here's your prize: my knight! 

The rest of the game requires no commentary. Even Deadlost was silent.

47...Kg4 48.Kf7 a4 49.Ke6 Kxh4 50.Kd6 Kg5 51.Kc7 Kf4 52.Kxb7 Ke3 53.Kb6 a3 54.bxa3 Kd2 55.c4 Kc3 56.c5 Kd4 57.c6 Kd5 58.c7 Kd6 59.c8=Q Ke5 60.Qd8 Ke6 61.a4 Kf7 62.Qd6 Kg7 63.a5 Kh7 64.a6 Kg8 65.Qd7 Kf8 66.a7 Kg8 67.a8=Q checkmate