This bug was exploited again just before 3pm yesterday on a number of stocks, including the following (for evidence, refer to the stock charts, stock history on api.howison.co.nz, and my previous explanations): I find it disappointing that this issue still hasn't even been acknowledged. Bugs like this are directly costing iPredict a lot of money (about $500 just this time), aby hmonkey - General
If you short a bundle of n stocks, the payout would be $n-1, not $1 (as all but one will close at $0). Suppose there's a bundle of 5 stocks, and you have one share of one of those stocks. You then short the whole bundle for $4, except you then own +1 share and -1 share of the same stock. However this stock closes, you'll make $1 (if it closes at $0, you make $1 from the short share, and if itby hmonkey - Feature Requests
mrh Wrote: ------------------------------------------------------- > Well, there's no trading on most state-level > stocks precisely because there's nothing really to > trade about. The US system is such that some > states are safe and only the battlegrounds are > interesting. > > But Florida has had over 600 trades today; with > that level of activity I think pby hmonkey - General
mrh Wrote: ------------------------------------------------------- > You can put in orders of your own, you know. And > on some of the US election markets these are now > the vast bulk of the liquidity. Yes, but in my experience people are considerably less likely to do that if there's no market maker, often leaving stocks with $0.9997 spreads. I suspect removing them will severeby hmonkey - General
admin - would it be possible to have thin market makers (ie one stock per price point, with the current density), rather than no market maker? This wouldn't come at much cost to iPredict, and in my experience it massively increases market liquidity (and enjoyment) - ie people seem more likely to trade and place their own orders if there's also a market maker in play. I've brought this up previby hmonkey - General
admin: POLL.TV3.2.UF should also close at 0. MVM_Group: In the active orders section of your profile it states that "trading on stocks is not generally suspended before or during the deciding event". The admin can't be expected to react instantaneously to every event.by hmonkey - Closure requests
admin, the market makers for the stocks I list in post #2 are still affected. Will these be fixed?by hmonkey - Bug Report
Just received an email from the programmer reporting that it's fixed now.by hmonkey - Bug Report
admin, any word on whether these will be bundled?by hmonkey - Active Stocks
Lanthanide Wrote: ------------------------------------------------------- > Might be worth pointing out this thread while > you're at it: > Already have doneby hmonkey - Bug Report
I've been in touch with the programmer who is now aware of this and working on it. Note that the other version of the API (/app.php?do=api...) is still working at the moment.by hmonkey - Bug Report
I've been put in direct contact with the programmer about this now. He'll be looking into it this weekend.by hmonkey - Bug Report
I just did some more testing that strongly suggests that this is the cause of the server problems. Here's what I did (all at about 5am New Zealand time, so there wouldn't have been many other active traders): I used two test contracts. A buggy contract (AUS.ABBOTT.DEP) which is affected by the above bug, and a non-buggy contract (91.JUN12.MID) which isn't. I then did the following series of reby hmonkey - Bug Report
For admin/other readers: further discussion about this was here. Now that there's a media release, on balance I think this should probably close at $1, but there is a judgement call to make regarding inflation. Perhaps future contracts like this should be more explicit about how this is measured.by hmonkey - Closure requests
It seems pretty clear from the announcement thread that this should close at $1 for Romney. These criteria have been known by everyone for two months. There was a point before the announcement where I was buying the 'Other' stock, but the clarifications ruled that out.by hmonkey - Closure requests
Here's a complete list of all affected stocks that have been posted about in the forums (thanks maxb and mrh). Probably still not exhaustive. Currently affected: FALK.ARGAMB.EXPEL FALK.ARGAMB.WITHDRAW FALK.UKAMB.EXPEL FALK.UKAMB.WITHDRAW SKYCITY.BANKS.RECUSE Previously affected: ACC.LEAK.COLLINS ACC.LEAK.JUDGE ACC.LEAK.PULLAR ACC.LEAK.WHALEOIL ACC.LEAK.SMITH ACC.LEAK.LUSKby hmonkey - Bug Report
Fair point - 4.1% sounds right then. Regardless of the exact percentage I think there are still valid arguments for both interpretations (more than inflation or not), so I'll await clarification from the admin. There's also the question about the lack of media release.by hmonkey - Active Stocks
Can these stocks be bundled? Thanks.by hmonkey - Active Stocks
Total tax on petrol is about 59c, which would make it 3.4% (the stock doesn't appear to be specific to the National Land Transport Fund portion of the excise). I guess it depends what timeframe you use for inflation (there's been 6.1% inflation since the last increase). Hence the need for clarification from the admin - it certainly seems to be within the range of an inflationary increase. Plus thby hmonkey - Active Stocks
Sorry, I wasn't clear that I was talking about the petrol stock rather than the tobacco one (which is what g_sure's post also refers to).by hmonkey - Active Stocks
admin, a couple of questions about the judging criteria: I can't see any media release about the increase on the beehive website. If one doesn't come come, does that mean it will close at $0? 2c corresponds to about a 3.4% increase. What counts as "above the rate of inflation" if there's no explicit mention in a future media release? 3.4% is less than inflation since the last increby hmonkey - Active Stocks
I'm pretty sure the other one was a deposit (by comparing ROIs across days at ). The net worth and ROI rankings both update at the same time, at midnight each day - neither is live.by hmonkey - General
Lanthanide Wrote: ------------------------------------------------------- > However I only see a very few trades that bought > it up back up to the $1 mark: 15:53:05 there is > one @ 0.9826 with many other simultaneous sales > down to 0.5826 or so 1 at a time. The next buy to > 0.9826 happens at 15:53:12, followed by many sales > in blocks of 10 down to 0.5000 or so. Tby hmonkey - General
You can see the trading history at which quite clearly confirms it (although the trades are out of order when done simultaneously). For example on the GDP.DEC12.GT05 stock, it shows that at 15:51:51 yesterday it was sold down from $0.5663 to $0 with one stock per price point (plus a few active orders), then seconds later it was bought back up to $0.2198 with 10 stocks per price point. A minute aby hmonkey - General
Looking at the trading history for those stocks, it was the same market maker bug that's affected some other stocks in the past - where an order for one stock is replaced with an order for ten stocks in the other direction. Looks like it cost iPredict about $1500 in market maker losses just from those contracts. I don't see much point in naming names, but it's not hard to figure out who it was.by hmonkey - General
Remaining orders available for one more day only (edit: no longer available)by hmonkey - Offers to Trade
It has just been fixed.by hmonkey - Bug Report
admin, will you be fixing the issue with this bundle I reported here?by hmonkey - Active Stocks