Forums

Players who start at the lowest rating, is that sandbagging?

Sort:
BlundasBlundasBlundas

I think I have made my points clear. I am not going to sit around wasting my time playing these lower than 1000 players for 2,403 games like this guy in the game below. I can see the site doesn't care about what I brought up.

David

"How does being deliberate or not change the conclusion?"

The question of deliberation comes in as to whether people are trying to sandbag ie. be rated lower than they should be. They are deliberately losing some games - presumably ones they don't care about - so that they can win games that they do care about. What distinguishes those games? Their level of care factor, which is why sandbaggers are normally throwing away random pairings and playing at their best in tournament games, which leads to them getting caught and banned. If they're just doing it in their random pool games, maybe you'll get lucky a time they choose to throw one of their games. The problem is that you don't feel as offended by that so you don't dig into as much - it's a confirmation bias, like how you notice the red traffic lights you get way more than the green.

If someone's not doing it deliberately, then it happens to be within their first 20-25 games and after that their rating will get up to the 1000 level anyway.

"The lower levels have basic themes they need to work on that higher rated players already know."

If the lower rated people did in fact work on those themes, they would no longer be rated so lowly. I just don't think many people at that level care about it as much as you seem to - no-one else has posted this sort of complaint before, so you're definitely an outlier.

"But what we can do is reward wins more"

That makes no sense, because it would lead to rating inflation all over the place. Or an 800 player getting lucky in 1 game, getting pushed up into 1000 immediately, and then beaten back down with their next losses. People would be more annoyed about that sort of fluctuation than they are about the current situation, which only you seem to care about.

David

I also note now that you've closed this latest account too, so clearly you have no interest in sticking around. Any new account is going to have some reliability issues with their rating until they've played some games and been around for awhile - rather than attributing that to some sort of conspiracy theory, just play the games, enjoy them, and after a little while it should work out. If you care about your rating as much as you seem to, it will get to a level where you're a bit more insulated from these sort of vagaries. You may in fact see more of those folks joining thinking that they're 1000 rated players and then get a rude awakening.

I also think he's going to be in for a surprise if he thinks lichess handles things any differently. Or about the responses he may or may not get about it if he posts in their forums lol

BigChessplayer665
David wrote:

I also now that you've closed this latest account too, so clearly you have no interest in sticking around. Any new account is going to have some reliability issues with their rating until they've played some games and been around for awhile - rather than attributing that to some sort of conspiracy theory, just play the games, enjoy them, and after a little while it should work out. If you care about your rating as much as you seem to, it will get to a level where you're a bit more insulated from these sort of vagaries. You may in fact see more of those folks joining thinking that they're 1000 rated players and then get a rude awakening.

Besides sandbaggers can be beaten actually most people play worse against beginners because they are expecting an easier game or out of panic and other things

David

He was basically sore that he was stuck a lower rating than he felt he should be and losing to people whom he felt should be above where his rating was as well. Happens a lot - the rating doesn't lie, though, not after 20-25 games

PositionalGambits

@David
@BigChessplayer665

"Their level of care factor, which is why sandbaggers are normally throwing away random pairings and playing at their best in tournament games"

That is not part of this thread. No one is talking about tournaments. All those games in question were based on random pairings. Be on topic without red herrings.


"The problem is that you don't feel as offended by that so you don't dig into as much - it's a confirmation bias, like how you notice the red traffic lights you get way more than the green."

You completely missed the analogy then. It was in reference to THE WEBSITE evaluating performance. This is not about the PLAYER getting a red/green game. You COMPLETLY missed the analogy.

As for confirmation bias, you later say we should play 20-25 games. It is safe to conclude bots are worse or a bit better initially than your first random human pairing. Over time, your argument poses that the strength will increase, if not already it will eventually surpass the bot at that level. That leaves the player's actual level of play. I would say it would then fall between the bot level and a human level at the same "level". By this, I mean let's say the bot is 1100, but it's not 1100 (maybe 900). Ok, so the human opponent at 1100 would actually be something like 1300 or higher. It's not any news that humans at a level are much stronger than the bot at the same level. The player who is trying to "settle in" and find their rating would be between the bot and the human that beats them.

Are you claiming that after 20-25 games, I will fall back down to 400 (per this account and the two others in theory)? I find that confirmation bias on your part. Your belief in this is faulty. At the lower levels I do not agree that lower level players who play regularly will fall back down to 400. Instead, their complaints are usually, "I reached ____" (some rating). They then complain they can't get higher. They don't complain they lost all rating points.

So, now we have a range. Some high point and then some drop. This is not "tilt" as it is labeled. This is the drop that would happen within the 25 games you are referring to. What I am saying is we should be allowed to play a bot and guess that drop point. If I want to start at 1100 with regular rating point gains (+5 to +10), I should be allowed to. I shouldn't be given hundred point hikes which then make games useless. I end up bouncing from easy wins I don't learn anything from, then I get paired with someone who demolishes me and I don't understand how I lost. If I play 1100 and play people with ratings within "arm's reach", then I can study the parts that are manageable to gain awareness from.


"If someone's not doing it deliberately, then it happens to be within their first 20-25 games and after that their rating will get up to the 1000 level anyway."

9 so far.


"I also note now that you've closed this latest account too, so clearly you have no interest in sticking around."

Is that the "depth" of your engine? You only think one layer into things?


"He was basically sore that he was stuck a lower rating than he felt he should be and losing to people whom he felt should be above where his rating was as well."

I am not opposed to being "stuck". You really do miss the mark don't you? You have no clue about what I was stating. My comment was on the MOVING UP in rating NOT GETING STUCK. In fact, I would love to start at 1100, get 10 points for a win, lose 10 points for a loss. No more points initially. Then as you gain 100 points in rating, it goes down to 5-8 so you have to be more consistent and have more wins. Instead, you are using confirmation bias to say, "Well if we give everyone 100s of points then it inflates everyone's rating." Yes, if you BELIEVE that is the method to award points then YES that would happen. What if you BELIEVED that people should earn their rating points? Then you don't get inflated points.

I would be much more in favor of playing 20-25 games if I knew 20-25 were not going to be completely USELESS.

You really need to think about this and not just reply like a shill for the site.

David

TLDR except for the fact that you're not interesting in playing the 20-25 games normally to get a proper rating but want the site to get it right straight away through some sort of magical thinking algorithm. Fact is, no-one seem to care about it except you.

PositionalGambits
David wrote:

TLDR except for the fact that you're not interesting in playing the 20-25 games normally to get a proper rating but want the site to get it right straight away through some sort of magical thinking algorithm. Fact is, no-one seem to care about it except you.

Which of the 9 games did I NOT play normally? And how do you know I won't play 11-16 more games?

"but want the site to get it right straight away through some sort of magical thinking algorithm. Fact is, no-one seem to care about it except you."

I don't care about that either. If you actually read the thread, not just this account, you would see I want that "magical thinking" to not be an algorithim. You select the rating, and then you build from it. 10 for a win, 10 for a loss. After you gain 100 points, you get 5-10 for your next win. This avoids inflation, and you play more meaningful games, and you don't have to wait for your 20th-25th game. If you chose 2000, and you were really 1200 for example, you wouldn't get a win until ~80 games later. No one is going to want to do that. Not even 1500 with ~40 games.

If we can select bots at 1200, why can't we stay around ~1200 without such ridiculous pairings based on a system that mispairs players?