Efficient Markets on Betfair
Don't recall anyone saying they are on an individual basis, though I have seen heaps of times people trotting out the line or words to the effect:
"makes a mockery of the idea that Betfair markets are efficient!"
I must be reading places different to yourself and Euler who seem to see a lot of people saying they are efficient on an individual basis.
Perhaps I'm misunderstanding exactly what people mean by 'efficiency' but I've seen posts like these:-PDC wrote: ↑Sat Jul 13, 2019 3:20 pmDon't recall anyone saying they are on an individual basis, though I have seen heaps of times people trotting out the line or words to the effect:
"makes a mockery of the idea that Betfair markets are efficient!"
I must be reading places different to yourself and Euler who seem to see a lot of people saying they are efficient on an individual basis.
Surely not much value to be had is a suggestion the prices are accurate, or perhaps LP is talking about automation from a statistical point of view?LinusP wrote: ↑Thu Jul 04, 2019 8:56 pmMy experience with horse racing markets:
Scheduled off time: Accurate, not much value to be had
Actual off time: Not that accurate, some value to be had (dependant on the delay)
BSP: Not accurate, value to be had
People may disagree and it is dependant on the odds/market but the above comes out in my data / pnl.
-
- Posts: 863
- Joined: Tue Dec 15, 2015 6:47 pm
I remember discussing this once on the BF forums, some people adamant that markets are entirely efficient for events such as horse racing (lol).
As pointed out by examples on here, the point is that on average they may be pretty accurate - but we don't bet or trade on average. I presume no-one here backs every single horse in every single race
In isolation, specific selections are very often not efficiently priced. This is even less efficient if you look across different times, as suggested previously in this thread.
The simple range of 'difference' for the same selection demonstrates markets cannot be efficient all of the time. Nothing significant changes (mostly at least) for a horse's chance in reality, yet it's price can often be vastly different as the chart example previous on here shows clearly.
As pointed out by examples on here, the point is that on average they may be pretty accurate - but we don't bet or trade on average. I presume no-one here backs every single horse in every single race
In isolation, specific selections are very often not efficiently priced. This is even less efficient if you look across different times, as suggested previously in this thread.
The simple range of 'difference' for the same selection demonstrates markets cannot be efficient all of the time. Nothing significant changes (mostly at least) for a horse's chance in reality, yet it's price can often be vastly different as the chart example previous on here shows clearly.
-
- Posts: 863
- Joined: Tue Dec 15, 2015 6:47 pm
An interesting addition to this discussion, and pertinent to this forum specifically.
I believe it is often traders who create inefficiency - they may be less concerned of course about what the actual result of an event is, merely concerned with where the prices are moving.
A heavy laying trader for example may push a price much bigger, forcing smaller traders to move also. This naturally can create a market fundamental inefficiency as it has nothing to do with the chance of the actual horse in the race.
So we (traders) may be the cause!
I believe it is often traders who create inefficiency - they may be less concerned of course about what the actual result of an event is, merely concerned with where the prices are moving.
A heavy laying trader for example may push a price much bigger, forcing smaller traders to move also. This naturally can create a market fundamental inefficiency as it has nothing to do with the chance of the actual horse in the race.
So we (traders) may be the cause!
Not sure I understand the obsession about market efficiency, doesn't seem like something that traders should be concerned with too much. The fact that the prices like to take the scenic route before eventually settling on a "fair" starting price should probably be the main concern, if you manage to sell some of this volatility you've probably done a good job as a trader. If you patiently wait for the market to do something borderline stupid and push the prices to ridiculous levels then the market will probably reward you for your patience, I don't think you can do much wrong with this approach.
-
- Posts: 863
- Joined: Tue Dec 15, 2015 6:47 pm
Kai - I think there's a fair few who use this forum that also include position trades/bets in their portfolio, as well as just trading.
-
- Posts: 3140
- Joined: Sun Jan 31, 2010 8:06 pm
I guess a lot depends on how you're defining an 'efficient' market. If you're hoping for the market to settle at a fair price surely that relies on an efficient market to get it to eventually settle at a fair price. If you're waiting for the market to do something stupid like a spike you'd certainly be hoping for an efficient market to bring it back into line so you can cash in.Kai wrote: ↑Wed Sep 18, 2019 10:53 amNot sure I understand the obsession about market efficiency, doesn't seem like something that traders should be concerned with too much. The fact that the prices like to take the scenic route before eventually settling on a "fair" starting price should probably be the main concern, if you manage to sell some of this volatility you've probably done a good job as a trader. If you patiently wait for the market to do something borderline stupid and push the prices to ridiculous levels then the market will probably reward you for your patience, I don't think you can do much wrong with this approach.
- ruthlessimon
- Posts: 2094
- Joined: Wed Mar 23, 2016 3:54 pm
What I find crazy - when we test consistent random swing strategies (entries & exits @ market, no stops & targets etc), the longterm result (after all that risk), is normally a loss of the spread + commission (i.e. each trade has an expectancy of around -1.5ticks).
It's perversely beautiful
-
- Posts: 3140
- Joined: Sun Jan 31, 2010 8:06 pm
I'd have thought to a stats guy that wouldn't be crazy but expected. As soon as you start averaging large volumes of data surely you'd come to the conclusion that you'd end up down by the overround. I'd expect the same to be true if we took the market at any timepoint when the overround is a reasonable amount.
-
- Posts: 207
- Joined: Sat Jun 16, 2018 7:57 pm
Then try a non-random strategy.
ruthlessimon wrote: ↑Wed Sep 18, 2019 6:33 pmWhat I find crazy - when we test consistent random swing strategies (entries & exits @ market, no stops & targets etc), the longterm result (after all that risk), is normally a loss of the spread + commission (i.e. each trade has an expectancy of around -1.5ticks).
It's perversely beautiful
- ruthlessimon
- Posts: 2094
- Joined: Wed Mar 23, 2016 3:54 pm
It’s a really interesting pointspreadbetting wrote: ↑Wed Sep 18, 2019 6:49 pmI'd expect the same to be true if we took the market at any timepoint when the overround is a reasonable amount.
The reason I say crazy, is because usually ‘commission’ is labelled as the main friction – but taking a price seems a far bigger hurdle – however it’s a difficult solve (unlike overround, (although we'll cripple sample size)) because invariably when we offer, we show our hand & the assumption we get filled on 100% of winners - nah aint gonna happen
I wonder how much it’d cost to bribe a bf programmer for enhanced queuing - faster, smarter, cheat
You do not have the required permissions to view the files attached to this post.