最短且仍然可读的解决方案:

static public int fightMath(int one, int two)
{
    if (one < 2 && two < 2) return 0;
    if (one > 1 && two > 1) return 3;
    int n = (one + two) % 2;
    return one < two ? 1 + n : 2 - n;
}

或者更短:

static public int fightMath(int one, int two)
{
    if (one / 2 == two / 2) return (one / 2) * 3;
    return 1 + (one + two + one / 2) % 2;
}

不包含任何“神奇”数字;) 希望能有所帮助。

您可以创建包含结果的矩阵

int[][] results = {{0, 0, 1, 2}, {0, 0, 2, 1},{2, 1, 3, 3},{2, 1, 3, 3}};

当你想要获得价值时，你就会使用

public int fightMath(int one, int two) {
  return this.results[one][two]; 
}

相反，你可以这样做

   public int fightMath(int one, int two) {
    return Calculate(one,two)

    }


    private int Calculate(int one,int two){

    if (one==0){
        if(two==0){
     //return value}
    }else if (one==1){
   // return value as per condtiion
    }

    }

我想到的第一件事基本上与Francisco Presencia给出的答案相同，但有所优化:

public int fightMath(int one, int two)
{
    switch (one*10 + two)
    {
    case  0:
    case  1:
    case 10:
    case 11:
        return 0;
    case  2:
    case 13:
    case 21:
    case 30:
        return 1;
    case  3:
    case 12:
    case 20:
    case 31:
        return 2;
    case 22:
    case 23:
    case 32:
    case 33:
        return 3;
    }
}

你可以进一步优化它，使最后的情况(3)为默认情况:

    //case 22:
    //case 23:
    //case 32:
    //case 33:
    default:
        return 3;

此方法的优点是，与其他一些建议的方法相比，更容易看到1和2的哪个值对应于哪个返回值。

如果你无法想出一个公式，你可以使用一个表格来描述有限的结果:

final int[][] result = new int[][] {
  { 0, 0, 1, 2 },
  { 0, 0, 2, 1 },
  { 2, 1, 3, 3 },
  { 1, 2, 3, 3 }
};
return result[one][two];

使用常量或枚举使代码更具可读性 尝试将代码拆分为更多的函数 试着利用问题的对称性

这里是一个建议，但在这里使用int型仍然有点难看:

static final int BLOCK_HIGH = 0;
static final int BLOCK_LOW = 1;
static final int ATTACK_HIGH = 2;
static final int ATTACK_LOW = 3;

public static int fightMath(int one, int two) {
    boolean player1Wins = handleAttack(one, two);
    boolean player2Wins = handleAttack(two, one);
    return encodeResult(player1Wins, player2Wins); 
}



private static boolean handleAttack(int one, int two) {
     return one == ATTACK_HIGH && two != BLOCK_HIGH
        || one == ATTACK_LOW && two != BLOCK_LOW
        || one == BLOCK_HIGH && two == ATTACK_HIGH
        || one == BLOCK_LOW && two == ATTACK_LOW;

}

private static int encodeResult(boolean player1Wins, boolean player2Wins) {
    return (player1Wins ? 1 : 0) + (player2Wins ? 2 : 0);
}

使用结构化类型作为输入和输出会更好。输入实际上有两个字段:位置和类型(阻挡或攻击)。输出也有两个字段:player1Wins和player2Wins。将其编码为单个整数会使代码更难阅读。

class PlayerMove {
    PlayerMovePosition pos;
    PlayerMoveType type;
}

enum PlayerMovePosition {
    HIGH,LOW
}

enum PlayerMoveType {
    BLOCK,ATTACK
}

class AttackResult {
    boolean player1Wins;
    boolean player2Wins;

    public AttackResult(boolean player1Wins, boolean player2Wins) {
        this.player1Wins = player1Wins;
        this.player2Wins = player2Wins;
    }
}

AttackResult fightMath(PlayerMove a, PlayerMove b) {
    return new AttackResult(isWinningMove(a, b), isWinningMove(b, a));
}

boolean isWinningMove(PlayerMove a, PlayerMove b) {
    return a.type == PlayerMoveType.ATTACK && !successfulBlock(b, a)
            || successfulBlock(a, b);
}

boolean successfulBlock(PlayerMove a, PlayerMove b) {
    return a.type == PlayerMoveType.BLOCK 
            && b.type == PlayerMoveType.ATTACK 
            && a.pos == b.pos;
}

不幸的是，Java并不擅长表达这类数据类型。