HANDICAP SHOGI IS a great way to learn shogi, for two reasons. First of all, as you learn the techniques needed to break through and win at each handicap, working from largest to smallest, you will learn most of the techniques needed to do well in the opening and middlegame of even-game shogi. You will learn the most important castles and how to attack them, and you will learn how to judge material sacrifices and how to evaluate positions.
I myself learned shogi primarily in this way. The strategies of even-game shogi are far too difficult for the amateur player to understand without the proper background afforded by handicap study.
The second reason to study handicap shogi is that it is widely played. Most clubs in Japan and in the United States stipulate that handicap games should be played in formal club play and in many tournaments whenever there is more than a single rank difference between the opponents.
Moreover, the Shoreikai (professional training organization) also follows this practice in all official games, though except for special events only the smallest handicap is used. There is also the tradition of an annual game between the Japanese champion (Meijin) and the Japanese Amateur champion, at bishop handicap.
THE BASIC RULES
When a handicap is given, the stronger player removes the specified pieces from the board, and they are retired from the game. The player with the extra piece(s) is called "Black" and the player without is "White". White always moves first in a handicap game, as if black's first move was the "capture" of the handicap pieces.
In the rare case of mutual entering kings (Jishogi), when the time comes to count points, white gets credit for the pieces removed at the start. In some events, it is stipulated that in case of a draw by repetition or a tied Jishogi count, white wins, but this rule should not be assumed.
Because the extra piece(s) for black are on their initial squares, which may not be ideal, the handicaps are not quite as large as they seem; the extra pieces though useful are also targets for white sometimes. Some people feel that to make the handicaps fully valid the extra piece(s) should be in hand rather than on the board, which would be much more difficult for white, but this is not the way it is done.
One must also keep in mind that white need only recover half the handicap to equalize the material, since he will get to use the captured material himself. For these reasons, even fairly competent amateurs need surprisingly large handicaps to defeat professionals (assuming the pro is trying his best to win, rather than merely teaching).
There is substantial literature in Japanese on handicap shogi, which I have been collecting for twenty years, and this series will attempt to cull the best lines from this literature, supplemented when necessary with private analysis by pros or amateur players of 5 Dan and above, including myself.
Although the series emphasizes how black should play the handicaps, I have not neglected advice for the player of the white pieces, as amateur players are often totally at a loss as to how to make the game a fight when they start out missing pieces.
The choice of lines for black is not so easy. Some books give the "best" lines, usually meaning the lines that win most quickly with perfect play. However these lines often score poorly in actual play, because they presume that Black will know and remember every variation perfectly, which is not realistic.
Other books give simplistic systems which are easy to remember, but which don't work well if white takes the best countermeasures. Another approach is to favor lines that are instructive. As much as possible, I have tried to strike a balance by recommending lines which are sound but not too difficult to learn and remember.
This series will consider eight handicaps, in this order :
Of these handicaps, six piece is rather popular for teaching novices, two piece is probably the most frequently played of all, and rook handicap and bishop handicap are also fairly popular. Five piece handicap is not played much any more, but since the gap between six piece and four piece handicap is huge I feel (along with some of the pros) that it is important and sometimes the only way to make a competitive game.
Some clubs have done away with lance handicap, because some amateurs lack the specialized knowledge needed to take advantage of it and because many amateurs feel that if the difference in strength is that small the weaker player might well win an even game, with just the tiny advantage of first move. Still, lance handicap remains in widespread use in many clubs and in the Shoreikai.
Eight piece handicap is often used with novices; Kazuharu Shoshi 6 Dan recommends P7f, P2f-2e, S3h-2g-2f-3e and then P2d. Three piece handicap (rook, bishop, and usually right lance) is rarely used now, because the gap between two piece and four is not so large as to require it. Black should probably use two piece strategy, with the extra advantage that white will be reluctant to use his right knight due to possible bishop promotion on 9c. Also if the long diagonal gets blocked black can play B9g without fearing an attack on the bishop by ... P9e. Right lance handicap was abolished nearly a century ago. Silver handicap (either one) is occasionally used to bridge the wide gap between lance and bishop handicaps, but lacks official recognition. Opening strategy is usually quite similar to even game strategy.
How differences in rank are related to the handicaps is a matter of some controversy. Most clubs use systems that only partially offset the difference in strength. A typical Japanese system might be to use just first move for 1 rank difference, lance for 2, bishop 3, rook 4, rook and lance 5, two piece 6-7, four piece 8-9. However in modern times the ranks have become somewhat compressed, with each rank representing a rather wide range of strength.
In my experience the rank differences that actually make for a fair game, assuming a 5 or 6 dan amateur is playing white, are lance = 1/2 rank, bishop 1 1/2, rook 2, rook and lance 3, rook and bishop 4, four piece 5, five piece (right) 6, five piece (left) 7, and six piece 8 ranks.
Professionals in training use a lance for two pro ranks, a bishop for five, and a rook for seven; the pro ranks are much narrower than the amateur ranks. A full-fledged pro should be able to give handicaps to amateurs as if he were ranked 7 Dan, assuming he is playing seriously, or as a 6 Dan if he is playing three games at once, as is the common practice. Note that the handicaps are not evenly spaced; the gap between no handicap (or first move) and lance, and the gap between bishop and rook, are smaller than the other gaps in my opinion.
Here is some general advice about handicap play. At high handicaps (four piece and above) it is generally best for black to attack as quickly as possible, not bothering to castle, since breakthrough can usually be achieved without giving white too much material in hand with which to counterattack. Of course castling cannot be bad; former Meijin Kato Hifumi once wrote that black can go into the fortress castle at any handicap and do well, but this does give white time to achieve an ideal disposition of his forces, and black may later be forced to show some defensive skill.
At any handicap including one or both lances it is usually wise to attack on the edge rather quickly, because if the attack is postponed fighting may erupt in the center, and the extra lance(s) will be irrelevant, or even a liability.
When only major pieces (rook and/or bishop) are removed it usually pays for black to exercise extreme patience, only attacking when the position cannot be further improved by quiet means. Normally white must wait for black to attack. The principal exception is bishop handicap, in which case white often attacks first because in the original position the bishop is a target for the rook, so white will often have time to castle and prepare an attack before black has time to reposition his bishop and do the same.
In the early stages of the game, which we are studying here, the most important factor in evaluating the position is the material balance, assuming reasonable king safety. Only towards the end, when direct attack on the kings becomes possible, can material be disregarded. Consequently we need a rough guide to evaluate material exchanges.
The scale I use is a compromise between the opinions of various pros. With a pawn as 1, lance = 4, knight 5, silver 7, gold 8, bishop 11, rook 13, promoted bishop 15, and promoted rook 17.
A promoted pawn is actually more valuable than a gold, because it unpromotes on capture, and so is worth 10, while the promoted lance or knight is worth 9. However, the exchange of a promoted pawn (or lance or knight) for a real gold is still extremely desirable, because you gain a gold while your opponent gains only a pawn. Numerically, you lose 2 (10-8) while your opponent loses 7 (8-1).
Promoting pawns and then exchanging the promoted pawns for real generals is probably the single most important technique you will learn from high handicap play. It is also important to note that although the value of pawns is rather low, it is rather important to have at least one pawn in hand if you are missing a pawn and very important if you are missing two. Therefore the early exchange of a knight for two pawns is often very good if the opponent cannot get a pawn in hand.
Regarding notation, I follow the conventions established by George Hodges in 1976 in the British magazine "Shogi", except for omitting hyphens between the piece symbol and the move.
The files are numbered from 1 to 9 starting on white's left, while the ranks are lettered from "a" to "i" from white's side.
The first letter of the piece (with "N" for knight) is followed by the destination square, unless the departure square must be given first to avoid ambiguity.
"x" indicates "captures"
"+" indicates "promotes" or "promoted"
"=" indicates "does not promote"
"*" indicates "drops"
A good move may be followed by "!", a bad one by "?"
When evaluating the final position of a line, since we are talking about handicap games the issue is not who has the advantage, but whether black's advantage has increased or decreased.
So a comment like "white is doing well" means that given the strength difference implied by the handicap, white will be more likely to win, even though he is still probably losing objectively.