Beginners who do that on purpose may exist, but more likely most of such beginners do it because they are too lazy to learn about ranks and will drop Go before stop being lazy. Also they may very underestimate difference of Power between ranks.
possible solutions:
-show VERY clear rank explanation to new users and VERY recommend to play weaker opponents for beginners
-make “beginner mode” where stronger opponents are hidden even in Custom
2 Likes
Really my point was that I didn’t pick this person out as some kind of example as to what’s wrong with the system.
It looks like this person was being picked to demonstrate a problem, which you have latched on to, and I’m trying to demonstrate why this example might not have been presented in the most transparent way and might not be the most relevant to demonstrate that beginners are having a bad time in general.
No matter what the matching system is, no many how many forms or checkboxes you fill out when you sign up, you can still choose opponents that you’re going to lose against, we can’t stop that and I don’t think that’s the issue.
1 Like
If they’re too lazy I’m not sure how you make it clear to people that won’t read things that 1d is super strong, 25k is beginner but still probably a good bit better than some people who’ve just learned the rules.
There’s a balance between making someone do a lot of things when they sign up, read this, learn about this ranking, read the rules, learn how to score and so and, and just letting them play games.
I think that’s probably the difference though between a ranked game mode and an unranked one. If you haven’t read the rules and you just start playing you’re gonna get stomped by 25kyus even. It doesn’t particularly matter who you’re paired with.
You could require that some people only play a subset of players, maybe even unranked, or you could
but what do you do when then they have no opponents, because another beginner isn’t available, maybe there’s no one online with timezones, maybe they’re playing other people their rank and not a new player, maybe other new players are being taught the game by higher ranked players, or maybe they’re just playing bots.
Do you just say “eh no-ones around, come back later”?
Maybe I’ll just make a new account and say no to beginner mode, because it’s hiding all of the actual games and challenges.
1 Like
I don’t see lack of transparency by filtering ranked games only.
Those unranked 10 games seems anecdotic experience. It shouldn’t hide the crushing experience he had with ranked ones.
1 Like
You pair with the nearest.
Besides it’s great to encourage those to come and not just consider our small family.
Sure and that’s still available with a 25k entry point.
I’m not against having beginner bots although they are yet not so interesting (well made). Goquest for example is populating a bit the poll this way.
The 3 early wins are what make the descent slow IMHO.
Can’t check what happened but seem unnatural considering @stone.defender evaluation of his level.
Note that if a stronger resign by being bored instead of being patient, the consequences are hard for the explorer.
2 Likes
This is the root of this problem.
If beginners play strong people, strong people resign.
How do we solve that?
3 Likes
There’s no technical fix for bad behaviour. People are allowed to cancel a game in the first few moves and it will be annulled so it doesn’t affect their rank. Past that point there isn’t really any excuse for not finishing the game. There used to be something in the TOS about playing to the best of your ability but it doesn’t seem to be there anymore. Perhaps it could be added back?
1 Like
I think that technical solutions can be put in place to work around bad behaviour that is expected and understood.
The excuse is that you can’t tell. The person is a [?].
Worse still, they may be showing as an SDK, if someone else already did this before you.
It’s only when their play makes it clear that “the game is a waste of your time” that the stronger person gets exasperated … often later in the game.
I can dream up various “things we can do” to mitigate against this, though none of my own solutions pass my own initial test of “yeah, nah, that’s not a lot better”…
3 Likes
Can we start by allowing beginners to play weak players if they want to? Like a ranked game against a 25k. Any more complete solution needs to at least have that be possible.
8 Likes
I see. Well, restricting people to 9x9 and / or live games until they’ve got rid of the [?] might be an improvement. Much shorter games, people less likely to get bored.
2 Likes
Are we stuck in a time loop?
3 Likes
Iterative exploration of concepts, narrowing in gradually on what might be possible 
3 Likes
Feel free to give your learned opinions from the many iterations of the time loop once again.
@GreenAsJade might be able to implement them.
1 Like
There are two “classes” of solution that we’ve been over:
- Ask the person if they are a beginner
- Don’t ask, just use the uncertainty
I think we need to know from @anoek whether solution type #1 is going to be acceptable.
2 Likes
I think this mischaracterizes the problem by framing it as a problem with strong player behavior.
They resign for the same reason you resign any game: because the time and effort is not worth the chances of winning, even if they seem guaranteed.
The only difference here is that the chance is high, but so is the time cost.
If winning were the only factor in resignation decisions, there would never be resignation, as any chance of winning is worth any amount of time, even if it’s microscopic chances based on the opponent suffering various incredibly terrible decisions.
So to solve the problem, you would need to somehow make the time cost worth the guaranteed win, or ban resignation in those situations.
If you make the game shorter the beginner becomes confused why they’ve already lost, and if you ban resignation the stronger player becomes frustrated at how much time they’re losing.
There’s really not a win here except in games that were already going to be short.
3 Likes
I think you misunderstood my intent.
When I said “how do we solve that” I meant “how do we avoid putting strong players in a position where they are playing weak players who are wasting their time”.
3 Likes
I expect this to be much less of an issue when players new to OGS are asked about their level and assigned a corresponding rating entry point, so weak players new to OGS are not paired as often against strong established players:
Or are you looking for a different solution (other than cancelling), which might work even if the system stays as it is?
1 Like
I imagine the new Anti-stalling feature will help a lot with the problem of “Win by Boredom”. Now the strong player can force a server decision.
However, I can’t emphasize enough that the effect of few erroneous wins is tiny compared to the effect of only playing strong players (previous comment). Here is a simulation I ran that compares a few scenarios of a new 25k:
- normal matchmaking (only play users at your current rank)
- only playing stronger players (5k)
- normal matchmaking with 3 “boredom wins” at the start
Clearly the boredom wins have an adverse effect, but, on average, the user will still pass 20k after about 15 games. On the other hand, the 25k who only plays 5k won’t pass 20k until about game 70, and won’t get to 25k for hundreds more.
The source for this simulation is here if anyone want’s to poke holes: goratings/test.py at master · benjaminpjones/goratings · GitHub
A couple of things to know about this simulation at a high level:
- Even for 25k vs 5k, we expect an occasional upset - about 1 in 200 games
- There is a lot of variability due to randomness, especially in the 5k scenario. I account for this by running the simulation 100 times, then averaging the graphs.
7 Likes
minor critique, but it gets the point across, is that one can’t play only 5k once they descend below 14k due to the nine stone diff rule for ranked games.
a feature that was implemented precisely to deal with this weakness in Glicko design
3 Likes