Hello, hope you are doing well. So, we are back again on Poly Market. I want to do some trading. I want to look up some strategies using codeex cloud code and you saw the title, right? And just to show you that it's not clickbait. You can kind of see the trade I did here, right? You can can we zoom in a bit? So, you can see the odds were 1%, right? We put one up, we went we 100, right? So, that's a 100x trade. So, this is a strategy that's a bit strange. So, I kind of wanted to kind of look a bit more into it today, maybe using like claw code and stuff and just to see what happens in these strange trades. You can see we have a bunch of these. We have this, for example, this one. This is kind of on the Bitcoin up and down. How do I get this? Okay, so you can see again this time we have the 2% there and we've been 50 from one, right? So, this is a bit of a strange trade. So, I'm going to try to look a bit into that today and kind of see how I did this. Okay. So, if we go to cloud code here now, you can kind of see uh this is just confirmed these trades, right? Just a proof of trade. So, this time we got the 48x here, right? We put up one, we got 59 back. And we can of course confirm this by going to kind of the poly scans here if you wanted to just to see that it's actually going through and it's just that not just like a visual bug. So you can see this was success and yeah you can go back and check this if you want to but basically the idea that I wanted to do in this video is basically just to look at these trades and to see exactly what is happening because they are very strange and this kind of like a low risk high value but again it's not really so it's a bit strange but it's really fun to use like claw code or like uh codeex to analyze these trades. and try to learn from them and make it into like a repeatable strategy. So I thought just talk a bit about that today and I'll show you my setup here. So the way I stumbled over this trade was I think it was just by accident. So I kind of wanted to dive more into it. So if we just copy all of this kind of this proof of trade we extracted and let's say you start up like a new codeex. Uh I want to do like dash dash yolo right for codeex. So I want to run anything like dangerously skip. So basically we can just paste in this now and I can start asking about it because codecs are really good or clo are really good at kind of looking at everything what happened in like a big overview because it's kind of taxing to kind of go through all of these uh poly scan and everything right so it's really nice way to use AI coding agents and stuff to kind of analyze this. So, let's just set it up here and try to make it understand what actually happened in this trade here. So, I'm just going to go I made a strange trade. I don't fully understand. Can you investigate, research, explain what happened, please? So, now I'm just going to put codeex on this task here and it's going to of course look at all the hashes here. It's going to check up uh the wallets and everything on this trade we did here. So, you can kind of see I did this back in like a couple of months ago now or like a month ago. So yeah, I'm not kind of spoiling any strategies here like uh there there's no there's no secrecy around this, but uh it's been a while since I actually went back to this now. So let's just check it up and let's see what we can figure out from this. Okay, so you can see now uh Codex kind of finished up here. You bought both sides. So but the combined trade was profitable. And then we're going to go through kind of what happened here. So the window ended at X this time. Your down by filled at 1 second after the window ended. Okay, that's cool. We paid $1. We received 50 down shares. Okay, that's standard, right? But down was the winner. So those shares later redeemed for $50. Okay, that is exactly what happened like how we actually earned the money. But we also bought up that field, right? We paid one, we received 50 up lost. So the shared page zero. So this is what happened, right? We actually managed to fill both legs. So we also we we filled um 50 at for one, right? But we also feed 50 for one uh both up and down. And this had to mean that we had to win, right? Since we had both sides and of course the redemp uh redemption showed that we redeemed this for 50. So why this looks strange? Both fields were maker fields from your wallet uh which poly market has a buy. So they were likely resetting or just posted orders from your bot account that you got matched around the close. So there is something strange here uh happening. So I'm going to ask Codex now how could we possibly fill both sides right 1 second after the window opened. Okay. So I guess uh it was 1 second after the window opened. Uh it was 1 second after the 5m minute window closed or gas fine. But basically the idea so the strange part is how did we fill after open? Why were the cheap O2 order still fillable after the outcome was basically known after the window ended? Most likely your bot had resting low bids on both sides and um another taker swept them during the close settlement delay. Okay, that's interesting. Is that something we can take advantage of around the close settlement delay? Right. One Phil hit the winning side at the stale price, the other hit the losing side at the stale price. Basically, if I say so, we had to win uh anyways, right? Because if we had both sides, we have to win, right? Yes. If both filled, you're guarantee positive payout. Yeah, of course. So now we can ask about some strategies to replicate this. Just ask like brainstorm or something. Cool. Brainstorm some strats to make this happen again please. Yeah. Yeah. It's fine. But uh this is something you can just do. So I wanted to kind of showcase today if you find something if you have an ID I really like to leverage like AI codeex cloud code and stuff like that. If you have data that's even better to try to analyze what happens. So now we're going to try to come up with some strategies how we can kind of try try to replicate this or give us the best chance to this to happen again. Right? So let's see what it comes up with here. The edge you want is get paired exposure where up cost and down cost co cost is below share. Ideally both sides at 1 cent 2 cent. Of course this has to be rare, right? Because if not everyone could just do this. And here it comes up with some strategies. Two-sided cheap maker bids. Okay. Post close winner side snipe. A hybrid a short a pair accounting. So, I'm not going to go specifically into all of these strategies, but again, today I just wanted to talk about how you can try to leverage these chat bots, chat bots, these uh agents like Cloud Code and Codeex, Open Code and stuff like that to try to come up with strategies you can test. At least this is something I find very funny. It's not something I plan to to live off or like make a lot of money on. I just think it's really fun to try to look up and find new strategies we can test out and see how well they work. So I'm going to take you now to a couple of strategies I have to try to replicate this and I'm going to talk a bit about bit more about them. First what I want to talk about is kind of looking at new windows because Poly Market of course has to create like a new window every like I think they do it like every 24 uh hours in advance. So if you kind of look at the windows here, this is the live window now 4:15, right? And we can look at the future windows. This is like 10 minutes, but we have more here. So we can also scroll down and you can see this is 7 hours. But I think if we look at kind of the the API and stuff, we can kind of go down to I think we can do like 24 hours uh or something in advance. I think you can do that. So you can find windows that open like 24 hours from now. And I wanted to see if we can do something around that. So if we go to yeah the terminal here. So you can see here we are placing orders uh 24 hours ahead. Okay. So right now we're looking at yeah this window and we have a couple of orders on this. So this strategy is target not targeting the current 5m minute window. It's uh catching newly listed future windows roughly 1 day early and then resting uh 1 cent 50 con shares bids on both sides and we also have a cancel timing on that right but basically this is the strategy I'm going to try out. So if you look at the dashboard for this, you can see we are up and running here, we have now put down 105 orders, 104 orders. And you can see we are kind of, if you go down here, you can see we are preloading and we are kind of looking at future orders. So we're going to place this like 24 hours in uh advanced you I guess you can say that. So that is one strategy I'm trying to do just to try to fill up a fresh window uh that we put down these resting contracts or shares or bids right on both sides both up and down. So I can't really conclude that in this video but I'm kind of if it works I'm going to do a follow-up. Okay. But basically that is the first strategy I want to try. And if you kind of think about it, it's more like here we have like a new window spawn and this is a fresh window, right? There are no orders or contracts or anything shares or bids here. So we want to place like a 01 bid as quick as possible into this new window because there might be some kind of Q system here, right? And the one that places this first uh is the first on the list here, right? So maybe if we are up here and someone tries to fill this, we get filled first or something. I'm not sure, but this is the strategy I wanted to look at. So we're just going to see how that turns out. We will just have to wait and see, I guess. So it's going to take some time. Not in this video. I'm not going to see how this works. But this is something I'm going to try. And I wanted just to do one more thing, one more strategy that could be a bit more um yeah, live I would say. So let's see uh see about that one. And so the next strategy uh I kind of went back and forth with codeex cloud code and we came up with a strategy that is going to try to time exact the window switch. So exactly when kind of this goes to zero. We're going to place uh on both sides like up and down. And we're not going to ride this throughout the full window. Uh, I'm going to explain why, but basically we're just going to after 120 seconds, we're going to cancel those because if we cancel those orders, we don't pay anything. So, we only pay if the order gets filled. So, I'm going to show you kind of why I don't I want to cancel these after 120. But, let me just show you how this uh works now. So, you can see now this window is ending, right? And now we're going to try to buy into a new one. And we did cross over. Let's see if we got any bids sent here. Yeah, you can see we placed eight bids. And this was exactly in the window switch, right? And you can see we have one BTC up and down, one ETH, Ether, Solana, and XRP. So, we try to fill everything. So, we cover four windows. And we're just going to let this ride for about two minutes. And then we're going to try to cancel again. So I'm going to explain why. So if we head over here, right? And let's say we have we have the window, right? So here's like zero and here's like 5 minutes and let's say 2 minutes is like here. So we wanted to buy in here, right? And we're going to keep that 01 bid running for like 2 minutes. And from here the bid could be like like 0.5 or something like that up here. But when you get closer to kind of the end of the window it could be something like 9 five and05 up and down. But if we have our bids here all the way to the end, we're just going to take been taking advantage of someone fills the this zero bid just to get some money back. So you kind of want to avoid that. So I thought we can just keep this in this range here. So after 2 minutes we cancel this. Let me see now. Yeah, we are kind of clo closing up now on 2 minutes. So sometimes we could get like a bad fill here. But sometimes we can get like a fill the market turns here and we kind of end up with a win anyway. But um I want to see if we can actually see this get cancelled now. If not, I'm just going to double check it in the in the terminal. Yeah, you can see I think I saw that got cancelled here because active went to zero. Yeah, that works. So, this is the other strategy I'm going to try for a while just to see what happens and kind of recent trades here. This is the last 12 fills. So if we get any fills, it will kind of show up here and you have like a fill rate. We will also have like a win rate. But uh again, I just made this video today. So I'm probably this is going to take time because this is a rare event, right? So probably I will have nothing for you in this video about the results. But if I make a new video, then it probably has been somewhat successful. Either this strategy or the other one. But the takeaway from this video is that it's super fun to use data you find ideas you have compare them using like cloud codec open code and try to come up with a lot of different strategies you can test out. So you can see I just found 10 here I can go through. So basically I just use this for fun and I learn about how market works. I learn about naked positions ups and down feeling both legs and everything. So for me this has just been learning about how market works because this is like a financial market right even though it's a bit strange it is kind of market you can learn things from even though it's just playing around with some small money I think it's more interesting than just reading it on paper actually doing it and testing it with real money I think it's more fun than just um uh yeah doing it on like paper and stuff like that I think I learn a bit more about paper money can be good of course but it doesn't really reflect the latency you get from like a real trade but uh this strategy at least is very low risk since we cancel the bids after 2 minutes and stuff like that and this strategy by prefilling orders in the future this is only get paid again if they get filled right and this also gets cancelled after two minutes I guess I didn't mention that. But yeah, that is what I wanted to talk about today. Basically, just having some fun playing around with Poly Market, learning about how this works. And if you want to know how to set up everything, I did a video a couple of videos ago about how to set up your uh Poly Market Trading bot if you want to go back and watch that one. So yeah, hope this gave you some inspiration. That is the idea behind these videos. is not to tell you exactly what to do, but basically give you some inspiration of strategies you can try out. And if you want to drop by uh our discord, you can do that and share your strategies. Okay, I wasn't logged in now, but basically we have a discord. I'm going to leave a link in the description to that if you want to drop by and have a chat about uh AI, automation, trading, everything around this topic if you're interested. So yeah, thank you for tuning in. Have a great day and I'll see you again soon.