Wicc
Auto-translated
Gentlemen, you have chosen the wrong method for detecting maphack. Analyzing two consecutive saves introduces too much distortion and a high probability of error. Situations like "hand slipped," "sudden adrenaline rush," "third eye opened," etc., are very real.
I personally observed many games from behind the backs of top Heroes players in Peter, and I saw even wilder actions there. Like a person taking 3 Dragon Utopias in different parts of the map in 1 turn, deftly using Dimension Door, Portal, and Fly. Just jumping at random and stumbling into Dragon Utopias.
When I played StarCraft, I spent a lot of time on the problem of detecting maphack there. And that task was also greatly complicated by the simply fantastic intuition of certain players.
So, one of the correct methods for detecting maphackers is, for example, the following: take several games of a person and review them day by day. Look for so-called "branching points," i.e., moments when a person doesn't roll according to a pre-planned strategy, but makes a decision that determines the direction of further actions, a choice from several roughly equivalent options.
Then, based on this branching statistics, conclusions can be drawn. Plus, make an adjustment for the player's overall skill. That is, top players have a higher percentage of "guesses" (i.e., correct decisions that do not directly follow from available facts or even contradict them) by themselves; that's normal, and it's partly what makes them top players.
So, maphackers almost never have episodes like "ran there, and there was nothing." That is, making a clearly disadvantageous decision in a situation where there are other more advantageous alternatives, equivalent with a closed map. A maphacker at a fork in the road will "at random" run to where something is.
And if a player constantly "fakes" here and there in small matters (and not only in small ones) tied to guesswork, while demonstrating strong and confident play in calculable situations, he is almost certainly clean.
I think most experienced players can recall situations where they frantically searched for control across the entire map, built a lookout tower looking for Dimension Door, and a Dragon Utopia with a blue beacon was nestled in a secluded corner of their home spawn. Maphackers don't have such episodes - some scout hero will definitely run past the Dragon Utopia, then call daddy. Either daddy will come himself "to scout the territory" or "maybe there's something there!"
That is, a maphacker's game is... what word to choose... polished, perhaps... a calm and confident realization of advantage, without flailing back and forth, contradictory decisions, etc.
A second, more cunning method of detecting maphack is tracking taken and untaken restarts. An honest player makes decisions based on what he sees at his position + what he sees in the tavern. But a maphacker also sees what the opponent has and what is on the map in general. Therefore, if the series of taken/untaken restarts poorly correlates with the objects visible to the player, but correlates well with the map in general, that is a reason to think about whether he is a maphacker.
And a third, completely cynical method is direct provocation. That is, in a real game, something like "scaring". Like standing with your main hero in front of a GO at the start of movement, then crashing into it and going about your business. The most suspicious scenario is if the opponent did not reveal the GO or your main hero, but still crashed his main hero home, bought troops, etc. A less clear one is if the opponent saw the moving main hero but did not react at all, because behind that hero there are no troops capable of breaking the GO. That is, a regular adequate reaction to a situation considering factors that the player formally does not see.
So the general idea is this: you need to look for episodes that exonerate the player. Some actions that a maphacker would be unlikely to take. If there are many such actions, then "unclear" decisions can be forgiven - humanity is strong precisely because it is diverse, thinks and makes decisions differently; considering another person's consciousness as identical to one's own is not correct.
Well, if the game is emphatically adequate, plus in places even "too adequate," conclusions can already be drawn.
BUT!
As many have already correctly noted - the main thing in this matter is the presumption of innocence + not washing dirty laundry in public. Catch a maphacker - don't play with him anymore. Forget about the rating points. The integrity of the community is more important. And every such scandal, regardless of the outcome, is a nail in the coffin. Veterans of the Chaos League, for example, fundamentally do not play online, precisely because they don't want to deal with maphack.
Although if the tournament is serious, you can assert your rights. But in any case, first with the player himself, then if there are no satisfactory explanations from him, with the judge, and only then make it public.
In StarCraft, for example, I did this several times: I catch a maphack, then write to the guy saying either you report this game as a win for me, or I go to the judge and you risk getting banned for the next few tournaments. Usually people preferred to get off with small losses.
P.S. The only vote for the "luck" option is mine. Based on the principled considerations outlined above.
P.P.S. I glanced at the saves briefly, did not draw any conclusions.
I personally observed many games from behind the backs of top Heroes players in Peter, and I saw even wilder actions there. Like a person taking 3 Dragon Utopias in different parts of the map in 1 turn, deftly using Dimension Door, Portal, and Fly. Just jumping at random and stumbling into Dragon Utopias.
When I played StarCraft, I spent a lot of time on the problem of detecting maphack there. And that task was also greatly complicated by the simply fantastic intuition of certain players.
So, one of the correct methods for detecting maphackers is, for example, the following: take several games of a person and review them day by day. Look for so-called "branching points," i.e., moments when a person doesn't roll according to a pre-planned strategy, but makes a decision that determines the direction of further actions, a choice from several roughly equivalent options.
Then, based on this branching statistics, conclusions can be drawn. Plus, make an adjustment for the player's overall skill. That is, top players have a higher percentage of "guesses" (i.e., correct decisions that do not directly follow from available facts or even contradict them) by themselves; that's normal, and it's partly what makes them top players.
So, maphackers almost never have episodes like "ran there, and there was nothing." That is, making a clearly disadvantageous decision in a situation where there are other more advantageous alternatives, equivalent with a closed map. A maphacker at a fork in the road will "at random" run to where something is.
And if a player constantly "fakes" here and there in small matters (and not only in small ones) tied to guesswork, while demonstrating strong and confident play in calculable situations, he is almost certainly clean.
I think most experienced players can recall situations where they frantically searched for control across the entire map, built a lookout tower looking for Dimension Door, and a Dragon Utopia with a blue beacon was nestled in a secluded corner of their home spawn. Maphackers don't have such episodes - some scout hero will definitely run past the Dragon Utopia, then call daddy. Either daddy will come himself "to scout the territory" or "maybe there's something there!"
That is, a maphacker's game is... what word to choose... polished, perhaps... a calm and confident realization of advantage, without flailing back and forth, contradictory decisions, etc.
A second, more cunning method of detecting maphack is tracking taken and untaken restarts. An honest player makes decisions based on what he sees at his position + what he sees in the tavern. But a maphacker also sees what the opponent has and what is on the map in general. Therefore, if the series of taken/untaken restarts poorly correlates with the objects visible to the player, but correlates well with the map in general, that is a reason to think about whether he is a maphacker.
And a third, completely cynical method is direct provocation. That is, in a real game, something like "scaring". Like standing with your main hero in front of a GO at the start of movement, then crashing into it and going about your business. The most suspicious scenario is if the opponent did not reveal the GO or your main hero, but still crashed his main hero home, bought troops, etc. A less clear one is if the opponent saw the moving main hero but did not react at all, because behind that hero there are no troops capable of breaking the GO. That is, a regular adequate reaction to a situation considering factors that the player formally does not see.
So the general idea is this: you need to look for episodes that exonerate the player. Some actions that a maphacker would be unlikely to take. If there are many such actions, then "unclear" decisions can be forgiven - humanity is strong precisely because it is diverse, thinks and makes decisions differently; considering another person's consciousness as identical to one's own is not correct.
Well, if the game is emphatically adequate, plus in places even "too adequate," conclusions can already be drawn.
BUT!
As many have already correctly noted - the main thing in this matter is the presumption of innocence + not washing dirty laundry in public. Catch a maphacker - don't play with him anymore. Forget about the rating points. The integrity of the community is more important. And every such scandal, regardless of the outcome, is a nail in the coffin. Veterans of the Chaos League, for example, fundamentally do not play online, precisely because they don't want to deal with maphack.
Although if the tournament is serious, you can assert your rights. But in any case, first with the player himself, then if there are no satisfactory explanations from him, with the judge, and only then make it public.
In StarCraft, for example, I did this several times: I catch a maphack, then write to the guy saying either you report this game as a win for me, or I go to the judge and you risk getting banned for the next few tournaments. Usually people preferred to get off with small losses.
P.S. The only vote for the "luck" option is mine. Based on the principled considerations outlined above.
P.P.S. I glanced at the saves briefly, did not draw any conclusions.
