Refocused

After "A Correction" concerning how to play against the line 1.e4 e5 2.Nc3 Nc6 3.Bc4 h6 4.Nf3 Nd4!? I decided that I needed to look further into the Jerome approach, 5.Bxf7+!?, recommended by Yury Bukayev, after all (despite the fact that I still thought 5.Nxe5!? Qg5!? to be playable for White).

In the meantime, however, another Jerome Gambit game from Bill Wall has arrived via email, and I just finished another two Jerome Gambit games in my Chess.com "Italian Game" thematic tournament, winning one and losing one, so the next few posts will focus on them as an interlude.

A Correction

In my email comments to Yury Bukayev about the line we were discussing, 1.e4 e5 2.Nc3 Nc6 3.Bc4 h6 4.Nf3 Nd4, 5.Bf7!?, I did not show much enthusiasm. In short order, I got a very bright email, enlightening me.

How do you do, dear Mr. Kennedy!

Dear Rick, thank you very much for your 2 letters! But I disagree with your appraisals of 5.Bxf7 and of 5.Nxe5. I suggest you discuss with me or publish (it will be better) my following analysis (5.Bf7! Kf7 6.Ne5 Ke6 7.Qh5+- ) and my words about 5.Ne5 Qg5-+:

1.e4 e5 2.Nc3 Nc6 3.Bc4 h6 (Vienna game: Max Lange variation) 4.Nf3 (Italian game/ Three Knights game) …Nd4?

Paragraph 1. 5.Bf7! Kf7 6.Ne5 Ke6 7.Qh5!? (White stands better.)

I)  7…Nc2 8.Kd1 Na1 9.Qf5 Kd6 (9…Ke7 10.Ng6 Kd6 11.Qd5 mate) 10.Nc4 Ke7 11.Nd5 Ke8 12.Qg6 mate;

II) 7…Qg5 8.Qf7 Kd6 [8…Ke5 9.d3 (with the idea 10.Bf4 mate) …Nc2 10.Kf1!?+-] 9.Qd5!? Ke7 10.Qd4 Qg2 11.Nd5 Kd8 12.Rf1 White stands better.

III) 7…Qf6 8.Ng6!? (White stands better.)

     A)    8…Nc2 (8…Rh7 9.Qd5 mate) 9.Kd1 Na1 10.Qd5 mate;

B)   8…Qg5 9.Nf8!? Ke7 10.Ng6 White stands better.
C)   8…c6 9.Qg4!? Kf7 10.Nh8 Ke8 11.0-0!? Nc2 12.Rb1 Nge7 13.e5 White stands better.

D)  8…d6 9.Nf8 (or 9.Nd5) …Qf8 (9…Ke7 10.Nd5!? Kf8 11.Nf6+-) 10.Qd5 White stands better.

E)  8…Ne7 9.Nd5 Qg5 (9…Qg6 10.Nf4+-; 9…Qf7 10.Ngf4+-) 10.Qg5!? White stands better.

IV) 7…g6 8.Qg6!? (White stands better.)

A)  8…Ke5 9.f4 Kf4 10.0-0+-;

B)  8…Nf6 9.Qf7 Ke5 10.f4 Kd6 (10…Kf4 11.0-0 Ke5 12.d3+) 11.e5 with the very strong attack;

C)  8…Qf6 9.Nd5!? Qg6 10.Ng6 (White stands better.) …Nc2 11.Kd1 Na1 12.Nc7 (or 12.Nh8) …Kf6 (12…Kd6 13.Na8 Rh7 14.Nf8 White stands better.) 13.Nh8 Rb8 14.b3 (with the idea 15.Bb2) +-

V) 7…Nf6 8.Qf7 Ke5 9.f4 Kd6 [9…Kf4 10.0-0 Ke5 11.d3 (with the idea 12.Bf4 mate) …g5!(11…Nc2 12.Bf4 Kd4 13.Qc4 mate) 12.Rf6!? c6 (12…Qf6 13.Qd5 mate; 12…Qe7 13.Rf5+-; 12…Bg7 13.Rg6!? White stands better.) 13.Be3 White stands better.] 10.e5 with the very strong attack.

Paragraph 2. 5.Ne5?! Qg5.

Probably, Black stands better both after 6.Nf7 and after 6.Bf7, because Black has a very strong attack in both cases: Qg2, Nf3, d5, Bg4.

Do you agree with me, dear Rick? I suggest you to discuss with me or to publish (it will be better) my analysis.

Best wishes! Yury V. Bukayev (“Bruno’s Chess Problem of the Day”)

Thinking again...

Despite my initial skepticism about Black's chances in the line Yury Bukayev recently asked me about, 1.e4 e5 2.Nc3 Nc6 3.Bc4 h6 4.Nf3 Nd4 (a line I once dismissively suggested transposed to "the Improved Blackburne Shilling Jerome Gambit" after 5.Bxf7+), further examination indicates that things can get pretty tricky for White, if he does not respond properly.

Certainly White can simply 5.0-0 and have the better game (because of his lead in development), but the question arises: does the addition of Nb1-c3 for the first player "detoxify" the dangerous-in-the-original-Blackburne-Shilling-Gambit move 5.Nxe5? Of course, Black will respond with 5...Qg5.

Here, as with the original BSG, the greedy 6.Nxf7 leads to all sorts of complications and inevitable pain for White: 6...Qxg2 7.Rf1 d6 and Black's threat of ...Bg4 means that the first player doesn't even have time to win the Rook, safely (check out 8.Nxh8 Bg4 9.f3 Be7! 10.Ng6 Bxf3 11.Nxe7 Bxd1 12.Nf5 Nxf5 13.Rxf5 Bxc2 as one example, where White does not get enough for his Queen). 

He can carry on with 8.Nd5 Bg4 9.Ne3 Bxd1 10.Nxg2 Bh5 11.Ne3 Bxf7 12.Bxf7+ Kxf7 where White has a pawn for his sacrificed piece. Clearly it is not a line to go into voluntarily, if there are alternatives!

There is also the defensive-minded 6.Ng4, which can lead to either messy or kempt positions where White has two pawns for his piece: 6...d5 7.Nxd5 Qxg4 8.0-0 (or the messy 8.Nxc7+ Kd8 9.Nxa8 Nxc2+ 10.Kf1 Nxa1 11.Bxf7 Kc8 when both a-Knights are likely to eventually expire) Qxd1 9.Rxd1 Bd6 10.Ne3 Be6.

My preference, until recently, has been 6.Bxf7+, since after 6...Kd8 White can play 7.Ng4 and answer 7...d5 with 8.Ne3, covering the White g-pawn.

Of course, not everyone will agree with me. In fact, in my next post I will share a very educational email from Yury, taking this discussion in a new direction!

What do you think?

About a month ago, Yury Bukayev asked me about the following line of play: 1.e4 e5 2.Nc3 Nc6 3.Bc4 h6 4.Nf3 Nd4.

It had a certain familiarity to it, especially since he suggested 5.Bxf7+!? as a response.

I told him that I would share the opening and my thoughts about it with Readers.

In the meantime, I went to the online ChessLab site and searched for games with the line. I found exactly zero.

Silly me, I was looking in the wrong place. A check of the 26,685 games in The Database showed me 10 examples (White won 9).

One was Wall - Surr, Chess.com, 2010 (1-0, 11), which has already appeared on this blog.

So, clearly, it was time for me to pay attention. First, a brisk walk-through.

1.e4 e5 2.Nc3 Nc6

Starting out as a Vienna Game.

3.Bc4 h6 4.Nf3

The game has transposed to a Semi-Italian Opening, i.e. 1.e4 e5 2.Nf3 Nc6 3.Bc4 h6 4.Nc3 (I usually play the alternative 4.0-0).

4...Nd4

Reaching the diagram above.

This Knight move is seen in the Blackburne Shilling Gambit: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4. The idea in the BSG is that White's greedy 4.Nxe5 can be met by 4...Qg5. 

Then the classic BSG punishment by Black follows 5.Nxf7 Qxg2 6.Rf1 Qxe4+ 7.Be2 Nf3#. The pluckier 5.Ng4 is met by 5...d5, with Black advantage. "Best" at this point for White might be 5.Bxf7+ followed by 6.0-0, When Black has the advantage of a piece for two pawns, but many Jerome Gambiteers would be used to that.

Of course, against the Blackburne Shilling Gambit, many Jerome Gambiteers would reply with 4.Bxf7+!? to begin with.

Can the Knight jump work for Black in the delayed form, however? What do you think?

I am thrilled!

Yuri Bukayev has a very funny sense of humor. He recently sent me this email:

I found super news for you! Thus, I found on WGM Olympic Champion 2012 N. Pogonina's site (section "Press", publication "Natalia Pogonina Interviewed by Chess Rex", 30.07.2012) the following piece:

Chess Rex: If in chess… all pawns became bishops… then what would be the problem in playing the game? Or who will get the advantage…White or Black? If you are playing such game as white, then what would be your first move and why?

Natalia: This “game” is a mate in 4: 1.Bxf7+ Kxf7 2.Bc4+ (or 2.Bh5+) Ke8 3.Qh5+ Bg6 4.Qxg6# There are a few sub-variations (exercise: find them), but it’s still a mate in 4.

Thus, dear Rick, the Jerome Attack is theoretically the strongest opening, a winning opening, if... 

Are you glad?

In my post "On the Other Hand..." the other day, I mentioned Abby Marshall's "The Openings Explained" column at ChessCafe, where she recently took on "The Two Knights Defense, Center Fork Trick" (1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4 5.Nxe4 d5)

I also alluded to Yury V. Bukayev's analysis as well, which I sent to Abby.

Ms. Marshall had selected and added my email to the bottom of her column as "pertinent response."

Rick Kennedy from the USA – Abby, I never miss your column at ChessCafe, and love the hard work you put into each one. Here's an interesting Fork Trick resource. Please keep up the great work.

On the Other Hand...

When I wander over to the ChessCafe website, I like to read the monthly column by Abby Marshall (USCF Candidate Master, currently rated 2192) "The Openings Explained". She's a hard worker, she's not afraid of playing gambits (the King's Gambit is her specialty) – and she used to be from my home town.

This week's topic is "The Two Knights Defense, Center Fork Trick" [1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4 5.Nxe4 d5], something that has been discussed on this blog a number of times before (like when you play 1.e4 e5 2.Nf3 Nc6 3.Bc4 looking for a Jerome, and your opponent plays 3...Nf6; can your 4.Nc3 get him to cough up 4...Bc5, so you can play 5.Bxf7+ ?). For just a sample of posts, try "Jerome Gambit vs Two Knights Defense (Part 3)" and "Further Explorations (Part 1 and Part 2)"

Don't forget to review Yury V. Bukayev's analysis as well, which I will be sending to Abby.

Reflectogen

Readers know that I have encouraged those interested in new cutting-edge chess opening theory, and chess improvement in general, to visit Bruno's Chess Problem of the Day (see "Bruno's Chess Opening Articles" on my list of links). 

For some educational examples, see: "A New Link", "Improve Your Chess", "Something New in Something Old", "Even in the Most Respectable of Settings" and "Update".

The latest entry concerns 1.e4 e5 2.d3 Bc5 3.Nc3 Nf6 4.Nf3! known as the Nf3-Bukayev gambit-reflectogen.

Check it out!

Tournament Update

I have started my first 4 games in the Italian Opening tournament at Chess.com, mentioned earlier. Since 3 games were with the Black pieces, there was no chance for a Jerome Gambit (at least by me, although I played 3...Bc5 each time, offering my opponents the opportunity) there. In the one game with the White pieces, I tried for an Italian Four Knights Jerome Gambit, but my opponent opted to use the "fork trick" (lucky for me I have Yury Bukayev's analysis to help me find my way, there). 


This reminds me of a story from the San Antonio tournament 40 years ago, as reported in Bent Larsen and David Levy's San Antonio, 1972 : Church's Fried Chicken, Inc. First International Chess Tournament. American master Ken Smith played his Smith–Morra Gambit (1.e4. c5 2.d4 cxd4 3.c3) against Grandmasters Larry Evans, Donald Byrne and Henrique Mecking, and lost each time. When Mario Campos Lopez faced Smith, though, he played the French Defence instead of the Sicilian, causing Larsen to comment in the tournament book "stronger is 1...c5 which wins a pawn."


I am going to have to start annotating 3...Nf6 in my games with "stronger is 3...Bc5, which wins two pieces"!


So far, I have won one game in the tournament, and in the rematch with my opponent I will have the White pieces, so I am hoping... (Oh, no, he just played 3...Nf6).

Trick Fork

On the way to reaching the Italian Four Knights Jerome Gambit, Black can always stop off and play the "fork trick", which gives him, according to common belief, an even or slightly better game.* Is this an improvement over defending against the Jerome, though, where Black technically has a won game?


In the following game, my opponent gets tricked up in the fork trick...


perrypawnpusher - Tinde

blitz, FICS, 2012


1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4

Played after some thought. Was he considering 4...Bc5 ?

5.Nxe4 d5 6.Bd3 dxe4 7.Bxe4 Bd7

The main line is 7... Bd6, after which it is helpful to be familiar with analysis by Yuri Bukayev. A few days later, I faced 7...Bc5.**

8.Bxc6

Better was the straight-forward 8.O-O Bd6 9. d4, but I had ideas along the e-file.

8... Bxc6 9.O-O 

Now White has nothing after 9... Bd6, but Black comes to that conclusion a move too soon and makes a mistake. Once a defender has played his defense, he sometimes stops defending...


9...Qe7 10.Nxe5 


Now Black's best is to acknowledge that he has troubles down the e-file, but still go out and meet them: 10...Qxe5 11.Re1 Be4 12.d4 (if 12.d3, then 12...O-O-O 13. Bd2 f5 14.dxe4 Bc5 give Black a lot of active play for the pawn, for example 15.c3 fxe4 16.Qg4+ Kb8 17.Be3, even) 12... Qd5 (or 12...Qe6 13.f3 O-O-O 14.fxe4 with an edge to White) 13.c4 Qe6 14.d5 Qe5 15.f3 f5 16.fxe4 Bc5+ 17.Kh1 O-O 18.exf5 Qxf5 and White is a bit better.


Instead, he marshalls on.


10...f6 11.Nxc6 bxc6 12.Re1 Black resigned.

This time there is no minor piece to intercede between the White Rook and the Black Queen.


*Those who have read Yuri Bukayev's article on the "fork trick" may find that White, actually, has an edge. 


**perrypawnpusher- adrienr, blitz, FICS, 2012: 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4 5.Nxe4 d5 6.Bd3 dxe4 7.Bxe4 Bc5 8.O-O O-O 9.Bxc6 bxc6 10.Nxe5 Qd5 11.Nf3 Bg4 12.d3 Rfe8 13.h3 Bh5 14.Bf4 Re6 15.Bxc7 Rg6 16.Kh2 Re8 17.Bg3 f5 18.Re1 Rf8 19.c4 Qf7 20.d4 f4 21.dxc5 fxg3+ 22.fxg3 Bxf3 23.gxf3 Qxf3 24.Qxf3 Rxf3 25.Rg1 Rf2+ 26.Rg2 Rgf6 27.Kg1 Rxg2+ 28.Kxg2 Kf7 29.Rf1 Rxf1 30.Kxf1 Ke6 31.Ke2 Kf5 32.Kf3 h6 33.b4 Ke5 34.Ke3 Black resigned [I know it's a footnote game, but it was pretty well-played -- Rick]

Update

Readers who enjoy visiting "Bruno's Chess Problem of the Day" (one of the sites this blog is linked to) and appreciate Yury V. Bukayev's opening analysis (see "Something New in Something Old" and "Even in the Most Respectable of Settings" for examples) are encouraged to check out his further examinations of the Nh4-Bukayev Gambit (earlier article is here).

The Cure

White starts off the following game in typical Giuoco Piano fashion, and he seems to be looking for a quiet game – until Black gives him the opportunity to become very loud.




shepherd - garouss
blitz, FICS, 2011

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.d3 h6

This position is at least as old as Home - Puller, Cambridge, 1860, where White continued in cautious fashion with 5.Be3 d6 6.h3 Qf6 7.a3 (1-0, 25).

5.Nc3 Na5

No doubt looking for more piano after 6.Bb3 Nxb3, but this move allows White to deliver some magna instead. 

6.Bxf7+

Another example of the "Jerome cure."

Sadly, for Black, Fritz suggests that the best now is to not take the piece, but continue with 6...Kf8 7.Nxe5 Bxf2+ 8.Kf1 (a bizarre example of what Yury V. Bukayev might call "the Jerome-Jerome exchange" declined) when 8...Bb6 9.Ng6+ Kxf7 10.Nxh8+ Kf8 gives White a pawn and the exchange advantage. 

6...Kxf7 7.Nxe5+ Ke8

"Best" for Black was 7...Ke6, when White has a strong attack after 8.Qg4+ Kxe5 9.Bf4+ although he lost his way, and, eventually, the game, in chumbo - KvanHouten, FICS, 2011 (0-1, 32).

8.Qh5+ Ke7 9.Ng6+ Kd6 10.Qd5 checkmate

Even in the Most Respectable of Settings

These days, the Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+), or one of its relatives, seems to show up all over the place, even in the most respectable of settings.

For example, consider the analysis of the Giuoco Piano from Yury V. Bukayev, at the Bruno's Chess Problem of the Day site, with a new, in-depth look at  1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.c3 Nf6 5.d3 d6 6.Bg5!?

There is a lot to look at in the article, and I recommend it to those who play, or play against, the Giuoco Piano.

As a side note, the author took time to point out to me the following line, which he referred to (after 6...Bd7) as "the Jerome-Jerome exchange": 7.Nbd2 Na5 8.Bxf7+ Kxf7 9.b4 Bxf2+ 10.Kxf2 Nc6, an extremely rare tactical combination, he mentioned.

(I suspect that the "Jerome-Jerome exchange" may be rare in serious or well-played chess, but I cannot help mentioning that there are about 550 games in The Database with Bxf7+ for White and ...Bxf2+ for Black. Still that is only about 2% of the games in The Database, so I guess that is pretty rare, even among Jerome Gambits.)

Something New in Something Old

There is a phrase in chess, "annotation by result", which refers to the practice of judging a move or a series of moves by the outcome of the game. Won game? Good move! Lost game? Bad move!

Strong, inquisitive and creative players work against this tendency and are often rewarded with new ideas and positive results over-the-board.

Consider Yury V. Bukayev, whose opening discoveries have been mentioned here before. Recently, Yury has been looking at the Fritz Variation in the Two Knights Defense.

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6  4.Ng5 d5 5.exd5 Nd4

To quote from "Having Fun with the Two Knights" from Chess Asia, Volume 11, Issue 3, 1995, by Bobby Ang,

By way of a short historical background, this is known as the Fritz/Schlechter Variation, used by Hans Berliner extensively in his rise to first place in the World Correspondence Championship. Indeed, it seems to be a lot more logical than the usual 5...Na5 which locks the knight out of play on the edge of the board.

Some of our readers might be wondering how this opening got its name. In fact, if you are in possession of the excellent book by Warren Goldman on Carl Schlechter which is a biography and a collection of most of the wins of "The Austrian Chess Wizard", you might have noticed that this variation does not even appear even once. Lest you think that he lost all his games with this line we hasten to note that this Defense was suggested by the German player Alexander Fritz to Schlechter who analysed it in Deutsche Schachzeitung in 1904, thus the name.

6.d6

This line has largely been dismissed by the sources that I consulted.

6...Qxd6 7.Bxf7+ Ke7 8.Bb3 Nxb3 9.axb3 h6 10.Nf3

Black has good play for his sacrificed pawn in Bogoljubov - Rubinstein, Stockholm, 1919. He enjoyed his "two Bishops" and transitioned to one of those Rooks-and-pawns engame that he was famous for winning. What else did Bogoljubov expect? seemed to be the concensus of the observers.

Yury, in an email he sent me, enthused 

I think this new gambit is a distant relative (!!) of the Classical Jerome Gambit. Thus, the difference of Black's and White's material in my gambit and in Classical Jerome Gambit is the same after the acceptance of these gambits; the initial position (3.Bc4) is the same; White plays Bc4xf7; White plays without the white-squared bishop in result; Black's king is on f7 a in variant of acceptance of gambits etc.

I think that the line is interesting enough that I would point it out, even if it were not "Jerome-ish", but I can't resist sharing an odd line from the Blackburne Shilling Jerome Gambit Declined: 1.e4 e5 2.Nf3 c6 3.Bc3 Nd4 4.Bxf7+ Ke7 (we have seen this before) 5.Ng5 (a bit unusual, but many people play the move in many Jerome variations) Nf6 6.Bb3 Nxb3 7.axb3 d5 8.exd5 h6 9.d6+ Qxd6 10.Nf3 and, indeed, we have reached that position from Bogoljubov - Rubinstein, above!

10...e4 11.Ng1

A gloomy retreat. An unkind annotator might say White is already lost.

An Italian correspondence game between Antritter and Balletti in 1969 introduced 11.Nh4, but White lost in 18 moves. A rather obscure game played in Tennessee in the United States, 15 years later, R. Carpenter - S. McGiffert, tried an improvement, but White lost in 13 moves.

Yet, Antritter and Carpenter were on the right trail.

I will leave it to readers to visit Yury's website and learn about the "Nh4-Bukayev-gambit" which gives White new hope! 

The Jerome-Kennedy Gambits!?

Wow.

The other day I received an email from Yury V. Bukayev, in Russia, suggesting the description "Jerome-Kennedy Gambit in different opening systems."

It was similar to the encouragement that Bill Wall made a while back, that we begin to talk about the "Jerome-Kennedy Gambit"  when we look at 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+, in the manner of the "Smith-Morra Gambit".

Thanks, guys.

For now, I'd like to stick with using Alonzo Wheeler Jerome's name for the gambit, as I further research his efforts: the earliest being analysis published in the 1874 issue of the Dubuque Chess Journal, and the latest (that I have found) a correspondence game against readers in the Literary Digest in 1900.

As I noted in my afterward to the posts on the Literary Digest game [1, 2, 3, 4], Mr. Jerome has had a hard time holding on to "his" opening: sources such as Cook's Synopsis of the Chess Openings (1882), The American Supplement (1884), and Freeborough and Rankin's Chess Openings Ancient and Modern, (1889) were happy to keep the name "Jerome Gambit", but identified the chief analyst of the opening as "Mr. S. A. Charles of Cincinnati, Ohio." Sic transit gloria mundi.

Plus,

However...

If my Jerome Gambit article ever appears [insert laugh track] in Kaissiber, or if I do succeed in completing a book on the Jerome Gambit and it's relatives; then, I'd consider adding my name...

graphic by Geoff Chandler

All Rook and Pawn Endings...

If you want to play the Jerome Gambit, you are sometimes going to face the Two Knights Defense. You can try the Italian Four Knights Game, but you have to be prepared for the "fork trick"...

perrypawnpusher - rupsi
blitz, FICS, 2011

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4

Sigh. No Jerome Gambit today.

Still, I have studied the "fork trick"...

5.Nxe4 d5 6.Bd3 dxe4 7.Bxe4 Bd6

8.Bxc6+

Recommended by Y. Bukayev.

8...bxc6 9.d4 Qe7

An interesting idea. After my response, Black could have tried 10...e4.

10.0-0 0-0 11.dxe5 Bxe5 12.Nxe5 Qxe5 13.Re1 Qb5

All-in-all, my opening preparation has done its job, as White has a small advantage (better pawn structure) – as opposed to the Jerome Gambit, where he has a lost game.

Why my opponent guided me away from the shoals and into a safe harbor is an interesting question. For the next few moves he concentrates on exchanging off one of his isolated double pawns.

14.b3 c5 15.Bb2 c4 16.Qd4

White's position doesn't have a lot of "bite", but it you give it some time...

If Black now defends the imperilled g7 square with 16...Qg5, White has 17.Re3, threatening Re3-g3.

16...f6 17.bxc4

A silly move, when 17.Ba3 was available, winning the exchange because of back rank mate threats. 

17...Qc6 18.Qd5+ Qxd5 19.cxd5

White has come out with an extra pawn.

Unfortunately, it was now my turn to play "routinely" (read: carelessly), and my opponent quickly recovered the material.

19...Ba6 20.Re6 Bb7 21.c4 Rab8 22.Rae1 Bxd5 23.cxd5 Rxb2

The position reminds me of the old saying that "all Rook and pawn endings are drawn."

Still, I had made "something" out of the game after the "fork trick" and I wanted to play on a bit longer to see if I could do the same here.

24.R6e2 Rfb8 25.f3 R2b5 26.Rd2 Rd8 27.Red1 Kf7 28.Kf2 Rb6 29.Ke3 Ke7 30.Ke4 Kd6

31.Rc1 Rd7 32.f4 Rb4+ 33.Kf3 Rf7 34.Rc6+ Kd7 35.Rdc2 Rb7 36.g4 g5 37.f5 Ke8 38.Re6+ Kd8 39.d6 cxd6 40.Rxd6+ Rbd7 41.Rcc6 Rxd6 42.Rxd6+ Ke7 43.Ra6 Kd8 44.a4 Rc7

A slip that costs a pawn.

45.Rxf6 Rc3+ 46.Kg2

White's King should go forward with 46.Ke4 to eventually help his advanced pawn. Failing this, Black's active Rook can cause mischief – and eventually score the draw.

46...Ra3 47.Ra6 Ra2+ 48.Kg3

Again, my King's choice to stay "at home" dooms the game to a draw. I needed to see that 48.Kf3 Rxh2 49.Rxa7 was advantageous for White.

48...Ra3+ 49.Kg2 Ra2+ 50.Kg3 Ra3+ 51.Kg2 Ra2+ Game drawn by repetition

Hats off to my opponent, whose active play at the end of the game helped him grab the half-point.

A New Link

At the request of Yury V. Bukayev,  I've added a new link, one that will take readers to "Bruno's Chess Problem of the Day", a site that has much more than just chess problems. As it says on the site

Welcome on Bruno's Chess Problem of The Day. As its name indicates it, you will find here, every day, a new chess problem. You will find also a big chess problems archive, chess articles (including ones with theoretical opening discoveries), chess news, banks of ChessMasters games in PGN (sorted by opening and by player's name) and of course The chess directory.

I have referred to some of the opening analysis in an earlier post here (see "Jerome Gambit vs Two Knights Defense (Part 3)") and I think that readers will be interested in a number of the Chess Articles along with the rest of the content of the site.