Polymarketanything more than 'someone is saying'. You just believe ppl saying stuff on the internet?
anything more than 'someone is saying'. You just believe ppl saying stuff on the internet?
funkyduck参与话题 5消息总数 37
该用户的聊天记录定位。
聊天记录
共 37 条，显示第 1-37 条Is your proof 'im not the only one saying this'? Or you have actual proof?
Is your proof 'im not the only one saying this'? Or you have actual proof?
do you have any proof of this?
do you have any proof of this?
ukmo tells you that max temp will be X. What is the probability of X? What's the probability of X-1?
ukmo tells you that max temp will be X. What is the probability of X? What's the probability of X-1?
(not excusing, I think they're extremely incompetent)
(not excusing, I think they're extremely incompetent)
the previous version of ghost orders have been around for months, and it was way more egregious
the previous version of ghost orders have been around for months, and it was way more egregious
according to my rebates, i've been getting approx same amount of fills
according to my rebates, i've been getting approx same amount of fills
anyone checked if ghost fills are fixed by the update?
anyone checked if ghost fills are fixed by the update?
is it just me or does anyone else think that expiration 'fix' is almost certainly not going to fix anything?
is it just me or does anyone else think that expiration 'fix' is almost certainly not going to fix anything?
hmm yeah that sucks, seems like someone figured out another polymarket loophole
hmm yeah that sucks, seems like someone figured out another polymarket loophole
well known address, or something random?
well known address, or something random?
I see. So it seems like someone is splitting their orders and is able to pull the rest on fill of the smaller size. So maybe a better name would be 'ghost liquidity'?
I see. So it seems like someone is splitting their orders and is able to pull the rest on fill of the smaller size. So maybe a better name would be 'ghost liquidity'?
on chain you do not see the transaction at all, or looks like it went through?
on chain you do not see the transaction at all, or looks like it went through?
I see, so this is all seen through UI, correct?
I see, so this is all seen through UI, correct?
do you see the cancellation on-chain or is it visible otherwise? In this example 'you' are taker or maker?
do you see the cancellation on-chain or is it visible otherwise? In this example 'you' are taker or maker?
how do you detect / what do you mean by ghost fills?
how do you detect / what do you mean by ghost fills?
IB has had climate markets forever, just zero liquidity...
IB has had climate markets forever, just zero liquidity...
this is a thing that people used to do on CME for example as well, look up 'CME canary orders' or something like that. You simply have small order further up the queue, and you are able to pull the rest once you get priority info about order being filled. The trick is that fill messages get time priority before the update gets sent out to the rest of the participants. All this to say that if it took CME forever to counteract, polymarket will probably never do that
this is a thing that people used to do on CME for example as well, look up 'CME canary orders' or something like that. You simply have small order further up the queue, and you are able to pull the rest once you get priority info about order being filled. The trick is that fill messages get time priority before the update gets sent out to the rest of the participants. All this to say that if it took CME forever to counteract, polymarket will probably never do that
how do you detect them?
how do you detect them?
but they were late sending rebates as well fwiw
but they were late sending rebates as well fwiw
ignore what I just said then
ignore what I just said then
oh my bad you talking about liquidity rewards
oh my bad you talking about liquidity rewards
the last three were not there a few hrs ago
the last three were not there a few hrs ago
I actually did
I actually did
I got mine a few hours ago
I got mine a few hours ago
same here
same here
what jump queue problem?
what jump queue problem?
yeah that's most dumb - if there was simply a pinned post of 'we are aware of these issues, please wait while we resolve', but most of the times they are super evasive and you never know if you're just shouting in a hole when reporting any issue
yeah that's most dumb - if there was simply a pinned post of 'we are aware of these issues, please wait while we resolve', but most of the times they are super evasive and you never know if you're just shouting in a hole when reporting any issue
can I DM you?
can I DM you?
ty
ty
hmmm
hmmm
I am missing liquidity rewards payouts for the past 2 days
I am missing liquidity rewards payouts for the past 2 days
anyone missing maker rebates for past 2 days?
anyone missing maker rebates for past 2 days?
anyone missing maker rebates for past 2 days?
anyone missing maker rebates for past 2 days?
anyone suddenly getting cloudflare blocked with a weird message in API?
anyone suddenly getting cloudflare blocked with a weird message in API?
that's my claude investigating bugs in production
that's my claude investigating bugs in production
Bug report: *Root cause: a tick-flip race on Polymarket's own side. When a market's price crosses the flip boundary, the CLOB announces 0.001 via WS tick_size_change; the engine adopts it within ~100ms and re-quotes fine-grid — but Polymarket's order gateway keeps enforcing 0.01 for another 5–60s, even interleaving accepts and rejects of identical fine-grid prices (singapore 13:27 SGT: rejects at :23/:25/:26/:34/:35 bracketing an accepted fine ask at :29). The discriminating evidence: decision logs show our accepted resting orders were exclusively 0.01-grid right up to each bin's flip minute (so the REST seed was correct all day), and our first accepted fine-grid order lands seconds after each reject. The old reject handler just logged and retried the same price every cycle. It's also still recurring — 4 more rejects at 06:09–06:20 while the agent worked.*
Bug report: *Root cause: a tick-flip race on Polymarket's own side. When a market's price crosses the flip boundary, the CLOB announces 0.001 via WS tick_size_change; the engine adopts it within ~100ms and re-quotes fine-grid — but Polymarket's order gateway keeps enforcing 0.01 for another 5–60s, even interleaving accepts and rejects of identical fine-grid prices (singapore 13:27 SGT: rejects at :23/:25/:26/:34/:35 bracketing an accepted fine ask at :29). The discriminating evidence: decision logs show our accepted resting orders were exclusively 0.01-grid right up to each bin's flip minute (so the REST seed was correct all day), and our first accepted fine-grid order lands seconds after each reject. The old reject handler just logged and retried the same price every cycle. It's also still recurring — 4 more rejects at 06:09–06:20 while the agent worked.*