King's Gambit, Vienna Gambit & 97 more! | All f4/f5 chess openings!

#gambits #countergambits #fpawn #resource #research 

What do the King's Gambit, Vienna Gambit, the Jaenisch Gambit, and the Grand Prix Attack have in common?

Well, they're all common and tricky openings that are potentially very winning. Many chess YouTubers have described them enthusiastically, me included! I've played them all! Thematically, each of these are an example of an early assertive f-pawn advance in the opening; f4 for White and f5 for Black. The committal and consequential impact this early f-pawn move defines and names each of these openings.

As a particular fan of the Vienna Gambit, I pondered the following question:

How many named chess openings are there that are defined by f4 or f5, and which of these are good?

I had a quick look on the internet, but I couldn't find a previous exploration or compiled list on this bit of chess trivia. So, a couple of weeks ago, I decided to answer this question myself! And it's so interesting! Included on this list are curiosities such as the, Coca-Cola Gambit, the Double Duck Formation, and the Krazy Kat, Paris Gambit!

Identifying the lines

Initially, this idea came to me as simply a fleeting moment of whimsy. Those of you who follow my channel will know that my favourite opening is the Vienna Gambit (Falkbeer), and I’ve recently been exploring the Vienna Gambit (Max Lange). I had previously played the Grand Prix Attack against the Sicilian and still regularly play the Jaenisch Gambit against the Ruy López Opening, both of which have “Vienna-ish” ideas, but for Black. While making a canvas poster (16” x 24”) on the King’s Pawn Opening (get one from my merch store USD$15), I discovered some interesting f-pawn openings that I wasn’t aware of before.

For instance, the Ponziani Opening: Ponziani Countergambit (1. e4 e5 2. Nf3 Nc6 3. c3 f5) immediately caught my eye. The Ponziani Opening is an opening that I occasionally face with the black pieces, and the ostensible "best" moves for Black often play into the White's superior experience of the tricky lines. Here was a potential solution that reverses the fight and according to the Lichess community database, was effective for Black at both lower- and higher-rated levels! Fascinating!

Afterwards, I tried using the chess.com analysis function to find more named f4/f5 openings. There seemed to be lurking everywhere even from the first couple of moves! After I found around a dozen, most of which I had never seen before, I realised that I needed a more systematic approach rather than "brute forcing" it on the analysis board.

I did a quick bit of googling, surely someone has had the same idea as me before? Now, perhaps I didn't use the right search terms, but I couldn't find a comprehensive list. There were a few blogs and articles that described the usual suspects, but not a single compendium that answered my question at the top of the page, "how many chess openings are there that defined by f4 or f5, and which of these are good?". So, I decided to do create this myself!

Firstly, this was possible due to the incredible list of all chess openings on Jimmy Vermeer's website here: https://jimmyvermeer.com/openings.html At the time writing, the page has 3,729 named chess openings, along with the opening moves, and ECO (Encyclopaedia of Chess Openings) code.

The text search function inside Chrome for identifying strings of text within a page is a little limiting, so I opened the page in Microsoft Word to allow for some more advanced search options. This allowed me to find the openings, though the process was somewhat tedious. The rules I set myself for the named opening line to be included in the list:

• The opening must be defined and named by White playing the move f4, with the pawn previously on f2, or;
• The opening must be defined and named by Black playing the move f5, with the pawn previously on f7;
• Only a single instance of a specific position will be included, i.e., the list excludes alternate names (e.g., Jaenisch Gambit and the Schliemann Defense in the Ruy López are synonyms for the same position), and excluding different move orders that transpose into the same position;
• If technically different positions are referred to by the same name (e.g., Vienna Gambit occurs in the Max Lange Defense and Falkbeer Variations of the Vienna Game), each of these are included separately as long as they meet the above criteria;
• An opening with f4 or f5 earlier in the line, but subsequently named due to another move (e.g., Vienna Game: Main Line is defined by 3... d5), will be excluded from the list.

I identified 99 openings according to the rules above for move orders 1 to 6. On move 7, White had a further four openings with 7. f4 (Benko Gambit Fully-Accepted: Central Storming Variation; Benoni Defense: Modern, Pawn Storm Variation; Sicilian Defense: Najdorf Variation; French Defense: Main Line, Steinitz Variation), while Black had no named openings defined by 7... f5. I've chosen to limit the list to the first six moves as the additional four for White aren't thematically very similar to the others.

Evaluating the openings

Evaluating the "goodness" of a chess opening depends on what we mean by "good" and thus, can be context-bound. To improve the usefulness of the resource, I've included:

• Engine evaluation - Stockfish 16 NNUE (through Scid vs PC client on Windows 11) at a depth of around 40.
• Win vs loss data data from the Lichess database of community games, including all time formats, accessed in early January 2024. This data was stratified into two categories - lower-rated (ELO 400-1800) and higher-rated (ELO 2000-2500) - using the web interface of the tool on the Lichess analysis board.
• A calculated odds ratio for each opening, for winning for the player initiating the f4/f5 move.
• Qualitative labels based on the odds ratios (see below)
• Colour coded cells to help quickly identify the especially promising lines

Now, I can already hear the criticism here... 😂 These qualitative labels are simply a way of interpreting the data based on voluminous real world data in the Lichess database of the odds of winning (or losing!) from a specific position. Any opening is playable for an expert in that opening as the experience and study results in an asymmetry in knowledge and skill in the specific context of the line. However, I would argue that some positions and openings are clearly better (or worse) than others, on average. This would relate to intrinsic asymmetries how difficult a position is to play for human minds.

The thresholds above are based simply as as way of providing an intuitive ranking. Conceptually, the further an odds ratio is from 1 (balanced odds of winning and losing in the data), the more unequal the result.

As you explore the resource, you'd discover that the win odds ratios often don't align with the engine evaluation. An insight from this is that for most recreational chess, engine evaluation metrics and hence, suggested moves, need to be interpreted critically. The most accurate move might not be the "best" move, insofar as most playable for a win by humans.

Overview of the findings!

Big picture overview - there are 99 openings, more for Black than White, but proportionately, more of White's openings are good.

White:

• has 41 named openings, starting with Bird's Opening of 1. f4
• of these 24 are highlighted as promising (at least one "good" win odds ratio or better)
• several others have win odds ratio of "playable okay" for both lower- and higher-rated players, meaning greater odds of winning than losing
• many of the openings in the earlier move numbers can be considered as a proactive and early aggressive push - a classic opening gambit trading a pawn for rapid development and activity
• many of the openings in the later move numbers are part of more structured attacks that take control of the centre

Black:

• has 58 named openings
• of these, only 14 are highlighted as promising (at least one "good" win odds ratio or better)
• there are a very early cluster that follow the prototype of the Dutch Defense of 1. d4 f5, and these seem different to conceptually different to the f5-based gambits and countergambits
• many of the Black lines can be framed (and some are named) as countergambits - the data suggests that many are potentially unsound
• where the line is a gambit that wins initiative they appear to do better - especially where White has played a relatively slower opening
• some of these openings may be relatively complex - higher-rated players seem to have a better win odds ratio than lower-rated players

Highlights for White

Firstly, we have what is probably commonly known and already expected. The King's Gambit (1. e4 e5 2. f4) is good (especially in lower-rated games), as well as the gambits in the Vienna Game (1. e4 e5 2. Nc3): the Omaha Gambit (2... d6 3. f4), Max Lange Vienna Gambit (2... Nc6 3. f4), Falkbeer Vienna Gambit (2... Nf6 3. f4).

Secondly, there is a very similar cluster against the Pirc Defense and Modern Defense, using a three pawn approach - d4, e4, f4, which in various lines has the label as the "Austrian" attack. The King's Indian Defense: Four Pawns Attack is very similar with the inclusion of c4. I'm not terribly good against these hypermodern openings so this might be a good approach for me!

And from the perspective of finding something for White to use against a Black defense, the following lines seem very promising, with good and very good win odd ratios at both lower- and higher-rated levels!

• French Defense: Tarrasch, Closed, Pawn Center Variation (1. e4 e6 2. d4 d5 3. Nd2 Nf6 4. e5 Nfd7 5. f4)

• Caro-Kann Defense: Classical, Maróczy Attack (1. e4 c6 2. d4 d5 3. Nc3 dxe4 4. Nxe4 Bf5 5. Ng3 Bg6 6. f4)

Highlights for Black

Unfortunately, the landscape is a little more bleak for Black, especially against the very commonly played openings by White. That said, there is the group effective f5 gambits against the King's Pawn Opening, King's Knight Variation (1. e4 e5 2. Nf3), which includes some of the most common openings of all, such as the Italian Game, Ruy López Opening, and Ponziani Opening.

• the aforementioned Ponziani Countergambit (2... d6 3. d4 f5)
• Italian Game: Rousseau Gambit (2... Nc6 3. Bc4 f5)
• Ruy López Opening: Jaenisch Gambit (2... Nc6 3. Bb5 f5)

There is, however, a very winning line (incredible win odds ratio of 1.55 in higher rated games!) against the Center Game that will likely be unexpected for White: the Center Game Accepted: Paulsen, l'Hermet Variation (1. e4 e5 2. d4 exd4 3. Qxd4 Nc6 4. Qe3 f5).

The Good, the Bad, and the Ugly

The Vienna Game include the best lines f4, and curiously, also f5 as well! With White, the perennial and my personal favourite, the Vienna Gambit (Falkbeer) has the highest win odds ratio of 1.43 for lower-rated games, while the Omaha Gambit has the highest win OR of 1.34 for higher-rated games.

With Black, I've previously described my approach AGAINST the Vienna Game, by luring White into the Main Line, and then playing the anti-Paulsen Attack with the Bardeleben Variation (1. e4 e5 2. Nc3 Nf6 3. f4 d5 4. fxe5 Nxe4 5. Qf3 f5).

At lower-rated games, it has a win OR of 1.31. For higher-rated games, the best f5 opening is curiously against another Paulsen Attack, the one in the Center Game we saw before with the l'Hermet Variation.

The worst f4 opening is appropriately the meme line, the Grob Opening: Double Grob, Coca-Cola Gambit (1. g4 g5 2. f4) where the opponent wins roughly 2:1. GLORIOUS!

And with the "ugly", I wanted to highlight as they seem weirdly successful. The first is the Amar Opening: Paris Gambit (1. Nh3 d5 2. g3 e5 3. f4) that starts with the completely unreasonable king's knight to the edge of the board on the first move! And despite this, for higher-rated games, White has a very favourable win odds ratio of 1.26!

And some odd themes...

Going through the list, there are a couple oddities in the names of these f4 and f5 openings! There is a weirdly avian theme:

• Bird's Opening
• Canard Variation
• Pelikan Variation
• Double Duck Formation

... and American one!

• Massachusetts Defense
• Colorado Countergambit
• Brooklyn Variation, Everglades Variation
• Omaha Gambit

Please make use of this resource (spreadsheet on Google Docs) and tell me what you've found and learnt!