If markets could be graded in terms of overall volatility (say, by comparing the current price with the opening price or the price x minutes ago), and tables created that rank markets in terms of volatility, then it would help traders to find markets that suit their trading style.
Jeff
Volatility index
Hi Peter
That would be interesting to know.
Incidentally, perhaps the best way of measuring volatility would be in terms of the changes in the implied probability in percentage terms, eg a horse that steamed from 2.0 (50% implied odds) to 1.82 (55%) would be classed as having shifted by 10%.
Jeff
That would be interesting to know.
Incidentally, perhaps the best way of measuring volatility would be in terms of the changes in the implied probability in percentage terms, eg a horse that steamed from 2.0 (50% implied odds) to 1.82 (55%) would be classed as having shifted by 10%.
Jeff
That's not necessarily volatile, Jeff
A horse moving from 2.00 to 1.82 over a 5min period can happen quite easily. My idea of a volatile market is when a horse will jump from 2.00 to 1.90 in 60 secs, then move back to 2.10 within the next 60secs (as Peter says, Wolverhampton is the prime example)
How that could be monitored or graded though is impossible, without it simply reacting to movements, rather than actually predicting them
A horse moving from 2.00 to 1.82 over a 5min period can happen quite easily. My idea of a volatile market is when a horse will jump from 2.00 to 1.90 in 60 secs, then move back to 2.10 within the next 60secs (as Peter says, Wolverhampton is the prime example)
How that could be monitored or graded though is impossible, without it simply reacting to movements, rather than actually predicting them
I have all this data. In Bet Angel, if you go to the market overview screen you will see a summary there. I view volatility by % IP.
In the market overview there is a risk meter this shows you the volatility on the favourite adjusted for the market type, but not many people uses it to my knowledge.
In the market overview there is a risk meter this shows you the volatility on the favourite adjusted for the market type, but not many people uses it to my knowledge.
Peter - Thanks for that.
Would there be any chance of making is to that markets from multiple events could be loaded simultaneously into Market View, such as all of the markets in Guardian? That way, all of the markets you are interested in scanning could be ordered.
Also, a column that gives us the difference between the high and the low (expressed in terms of implied probability) would be useful, as it would highlight markets where there has been a lot of movement and those where there's been very little movement.
Finally, I think it would be good if 'high' and 'low'
could be user-defined in terms of timeframes, as otherwise distortions can be created due to someone getting a tenner matched at a ridiculous price soon after the market opened.
Le Tiss - I'm not saying that the hypothetical example I gave represents a large swing - merely that it illustrates a way of representing volatility.
What I'm proposing isn't a perfect solution. For example, a market could have gone all over the place but happen to be at the same place as it started out when you compile the table. However, I think it would serve as a handy filter that would highlight markets to have a look at. For example, if you specialise in high volatility markets, you might be interested in knowing about a market that was at 2.0 five minutes ago, but is now at 2.5.
Jeff
Would there be any chance of making is to that markets from multiple events could be loaded simultaneously into Market View, such as all of the markets in Guardian? That way, all of the markets you are interested in scanning could be ordered.
Also, a column that gives us the difference between the high and the low (expressed in terms of implied probability) would be useful, as it would highlight markets where there has been a lot of movement and those where there's been very little movement.
Finally, I think it would be good if 'high' and 'low'
could be user-defined in terms of timeframes, as otherwise distortions can be created due to someone getting a tenner matched at a ridiculous price soon after the market opened.
Le Tiss - I'm not saying that the hypothetical example I gave represents a large swing - merely that it illustrates a way of representing volatility.
What I'm proposing isn't a perfect solution. For example, a market could have gone all over the place but happen to be at the same place as it started out when you compile the table. However, I think it would serve as a handy filter that would highlight markets to have a look at. For example, if you specialise in high volatility markets, you might be interested in knowing about a market that was at 2.0 five minutes ago, but is now at 2.5.
Jeff
PeterEuler wrote:I have all this data. In Bet Angel, if you go to the market overview screen you will see a summary there. I view volatility by % IP.
In the market overview there is a risk meter this shows you the volatility on the favourite adjusted for the market type, but not many people uses it to my knowledge.
I thought that was just the Implied Probability as a function of the odds? Can you explain a little how it is adjusted for the market type?
Thanks
Peter
PeterLe wrote:I wonder if someone like James1st would have a view on this?
James - I appreciate that there are many factors in what you do, but amongst them, do you find you are more profitable on certain courses?
Sorry Peter I don't have any usable records of volatility by course or race type but certainly these 2 factors tend to demonstrate extremes.
I usually find that volatility is almost always a manifestation of thin markets although I wouldn't count Wolverhampton as a notable example; NH Flat races at Listowel would probably top my list.
My own view of volatility, in the context of my own trading, is whether I can trade the current race and for me that simply means that the odds aren't jumping around too quickly. There is little point in measuring movement in minutes, so what I would like to see is;
"how many odds points are hit in the last 5-10 seconds"
A low value of 2/3 is ideal for scalpers whilst values up to 6/7 is OK for trend or swing trading. Values in double figures (where the odds are simply unstable) would only suit those who are operating strategies where volatility either doesn't matter or is a positive advantage.
Thanks James
I think that sums it up nicely. Its about either choosing the markets you are going to trade that suits your own particular strategy, or if you have multiple strategies, changing them to suit the race in question. One size doesn't fit all.
Funnily enough; I was testing a system this afternoon (in practice mode) using 4 figure sums. I think i would have passed out if this had happened whilst i was trading for real. A very timely reminder. Always have your Betdaq window open too!
Sorry to see some folk have lost out. It really isn't good enough by Betfair in this age of high availability systems. I work in IT and there are many many organisations that strive for 5 x 9's availability, it just seems that betfair isn't one of them
I think that sums it up nicely. Its about either choosing the markets you are going to trade that suits your own particular strategy, or if you have multiple strategies, changing them to suit the race in question. One size doesn't fit all.
Funnily enough; I was testing a system this afternoon (in practice mode) using 4 figure sums. I think i would have passed out if this had happened whilst i was trading for real. A very timely reminder. Always have your Betdaq window open too!
Sorry to see some folk have lost out. It really isn't good enough by Betfair in this age of high availability systems. I work in IT and there are many many organisations that strive for 5 x 9's availability, it just seems that betfair isn't one of them
