- Blind mode tutorial
lichess.org
Donate

CHANCES OF WINNING BASED ON POINT DIFFERENCE

#1

Here is a chart that I made up that accurately gives you the odds of winning or tying against an opponent based on their performance rating.

Point difference
0- 50%-50%
100-40%-60%
200-30%-70%
300-20%-80%
400-10%-90%

So for example a 1500 player would have a 30% chance of winning or tying a 1700 player and a 70% chance of winning or tying a 1300 player. You can adjust the numbers depending on your rating or how you really feel your chances are but I would say this chart is a pretty good indicator of how you would do against different rated players.

Here is a chart that I made up that accurately gives you the odds of winning or tying against an opponent based on their performance rating. Point difference 0- 50%-50% 100-40%-60% 200-30%-70% 300-20%-80% 400-10%-90% So for example a 1500 player would have a 30% chance of winning or tying a 1700 player and a 70% chance of winning or tying a 1300 player. You can adjust the numbers depending on your rating or how you really feel your chances are but I would say this chart is a pretty good indicator of how you would do against different rated players.
#3

@BOBMILIN can you explain how you made this chart ?

Im interesting in your calculations to get those results ... (don't copy anything from here https://en.m.wikipedia.org/wiki/Elo_rating_system )

@BOBMILIN can you explain how you made this chart ? Im interesting in your calculations to get those results ... (don't copy anything from here https://en.m.wikipedia.org/wiki/Elo_rating_system )
#4

Mr Elo, get out of this body !

Mr Elo, get out of this body !
#5

I made the chart based on my experience in playing chess the last 15 years. I am not saying its 100% accurate but it could be used as a guide. All I know is that I rate around 1500 in classical chess and if I were to play a 1900 player I could easily lose 10 games in a row to a player with that rating and I would only expect to win or tie just 1 out of 10 times if that.

Now if I play a player that is at 1500 like me I would expect to win or tie at least 50% of the time so going from 1500 to 1900 I would go from 50% down to 10% in my chances so my chances would decrease as I played higher rated players ending up at 0% chance once I played players rated 400 or more points higher.

The best way to tweak this chart and make it more accurate is to play 10 games against players rated the same and then from 100 to 400 points higher and see how you do as you play each level.

I made the chart based on my experience in playing chess the last 15 years. I am not saying its 100% accurate but it could be used as a guide. All I know is that I rate around 1500 in classical chess and if I were to play a 1900 player I could easily lose 10 games in a row to a player with that rating and I would only expect to win or tie just 1 out of 10 times if that. Now if I play a player that is at 1500 like me I would expect to win or tie at least 50% of the time so going from 1500 to 1900 I would go from 50% down to 10% in my chances so my chances would decrease as I played higher rated players ending up at 0% chance once I played players rated 400 or more points higher. The best way to tweak this chart and make it more accurate is to play 10 games against players rated the same and then from 100 to 400 points higher and see how you do as you play each level.
#6

The numbers you're using aren't exact, but what you're describing is basically what the ELO system is designed to capture.

What's remarkable IMHO is how well this system seems to actually work.

The numbers you're using aren't exact, but what you're describing is basically what the ELO system is designed to capture. What's remarkable IMHO is how well this system seems to actually work.
#9

I'm unsure about this, to say the least.

If two super GMs play each other, the chances are they will most likely draw - even if there's 200 point difference between them.

If patzers like me play someone who's 200 points above me, I'll probably lose.

The likelihood of draws increases the stronger the two players are.

I'm unsure about this, to say the least. If two super GMs play each other, the chances are they will most likely draw - even if there's 200 point difference between them. If patzers like me play someone who's 200 points above me, I'll probably lose. The likelihood of draws increases the stronger the two players are.
#10

@Tim_Nice_But_Dim wrong my friend.

The elo system is'made in a way that 200 points difference will always results in the same win probability regardless of the ranking of the 2 players.

@Tim_Nice_But_Dim wrong my friend. The elo system is'made in a way that 200 points difference will always results in the same win probability regardless of the ranking of the 2 players.

This topic has been archived and can no longer be replied to.