Good Bishop vs Bad Bishop

A bishop's quality is decided by a single structural fact: whether your own pawns sit mostly on the same color square as the bishop or mostly on the opposite color. A bishop sharing its color with its own pawns is boxed in by them — every pawn on that color is a piece the bishop can never move past, since it's blocked by its own side's fixed structure rather than a moving target it could eventually maneuver around. A bishop on the opposite color from its own pawns has the run of the board those pawns don't occupy, which is exactly why the same-looking piece can be excellent in one structure and nearly useless in another.

Reading it off the pawn structure, not the current position

The fastest way to judge a bishop's long-term quality is to look at where the central pawns are fixed, not at how many squares the bishop happens to control this move. A pawn chain locked on light squares makes a light-squared bishop bad regardless of how open its diagonal looks right now, because as the game continues and more pawns settle onto that color, the bishop's available squares only shrink further. A dark-squared bishop in that same structure, by contrast, tends to improve over time as the position develops, since the light-square pawn chain never gets in its way.

A bad bishop is sometimes worth keeping anyway

A structurally bad bishop still has one job a knight often can't do as well: defending the squares of its own color, particularly around a king that's castled toward that side. Trading off a bad bishop for the opponent's active knight can look like an obvious fix, but if it removes the only piece capable of covering a critical color complex near your king, the trade can create a bigger problem than the one it solved. Judge a bad bishop by what it's currently defending, not only by how few squares it can attack.

The fix that actually repairs a bad bishop

A bishop that's bad because of a fixed pawn chain can genuinely become good again if the pawns on its own color get traded off or advance past the bishop's diagonal entirely, opening real lines for it. This is a real, plannable structural goal rather than wishful thinking — deliberately trading a same-colored pawn, or timing a pawn break specifically to clear the bishop's diagonal, is a legitimate multi-move plan on its own, distinct from anything else happening in the position.

"Good" and "bad" describe the structure, not permanent piece value

Neither label is a life sentence, because pawn structures change constantly through captures and advances. A bishop that's currently excellent can become boxed in a dozen moves later once new pawns lock onto its color, and a bad bishop from the opening can turn into the best piece on the board once the structure that trapped it dissolves. Reassess a bishop's quality every time the pawn structure changes meaningfully, rather than fixing a verdict on it early in the game and carrying that label forward without checking it.

For the separate question of holding both bishops at once rather than judging one bishop's quality, see bishop pair advantage; when the weak squares involved span the whole board rather than one bishop's diagonal, see color complex weaknesses.