Creating Tip Output to be read by a Bot

Mick, hiya.

This should probs have gone in the General fora, but no real matter.

From my perspective, I find using the WSDL at betfair incredibly easy - its the business logic at my end I have trouble with. So, my opinion on this would be for you guys to supply an API of some sort, similar to the betfair API. That sounds like a huge task, but if you have a look (this is a java based example, but I'm fairly sure it can be just as simple in other languages) it doesn't have to be as painful as it sounds. Obviously you would have to think about the architecture of the system, but its certainly doable, if you do things one step at a time.

I can't say I'm an expert on horses - I tend to find it easier to pick a broken nose than a winning horse, but from my perspective, what you would want to supply bot creators with is everything you have. Obviously, that can be a bit of a scary prospect, but you will in all likelyhood find that most bot users will only use a small subset of the calls you make available to them. A lot may end up using the others as well, because thats what bots and systems are all about, isn't it - finding an edge and using the data available to you.

From there, I think you could do a couple of things. One, you could create a framework with public methods which allow bot writers to dabble in it or you could allow bot writers full access to the API. If you did the former, you could probably speak to betfair about a licence to use the transactional API along with your framework. This would mean you could gain extra income from encouraging your punters to join betfair - a lot may already be members, but some won't be and income isn't really anything to be sniffed at.

If you did the latter though, you could still get the benefit of encouraging others to join betfair, but you could offer a similarly structured (if not as expensive) system of access to the WSDL.

So there you have it, in my long winded opinion, you should publish it as a web service with different levels of access.

What I want though is something like that, for football - I would only want to update the scores every ten seconds or so...

Personally, i'm a believer in Keep It Simple and would therefore put the tips into a simple delimited text format. As long as the format is consistent, it doesn't matter if each tip is separated by a line return or uses special characters such as the tilde "~" or goes for a JSON structure. Point is, it's dead easy to web scrape in any language, so anyone wishing to use your tips in a bot should be more than capable of downloading and parsing a few lines of text...

Thanks Guys

That is useful feedback

summary so far is an api that outputs KISS style information.

I guess I can see advantages in the api.

eg access can be restricted on a user by user basis.
someone with a malformed bot trying to pull info multiples times per second and possibly detrimenting the experience of others could be refused access.
Similarly we may have free and premium tips with access rights on a username basis.

Guess if the info is being built from a  back end db it would be relatively easy to provide more than one api.

One could output csv  another XML

Rambling now but possibly what the world needs is a standardised xml format for betting tips. :)

Picture how easy the world would be for bot makers if every service in the land who wanted to make bot readable info available produced it in the same XML format. ( perhaps with matching csv output .. the data after the 5th comma is always runner name for example. )

It would cut out the need to translate each and every different services output to suit your own bot.
Get every tip provider speaking a common language.
It could start here with you guys, move to simple dedicated little site with full info, gain momentum and live for ever :)

Guess its just really a matter of pondering over data to be transmitted and agreeing placeholder names.

( no need for every service to fill all place holders with info .. just what's needed )

As for the football Inksmithy  .. I also happen to run a tip site with bloke who as his day job works as an odds compiler for a major uk bookmaker. Issues with that at present may be with some of his bet types not being 100% suitable for exchanges. ( some pretty illiquid markets further down the leagues / the odd asian handicap bet at a bookie perhaps with no direct tie up to betfair asian quotes etc  )
Obvious possibilities however to develop there a more exchange only bot friendly type service.

All the best
Mick

Rambling now but possibly what the world needs is a standardised xml format for betting tips. :)
Picture how easy the world would be for bot makers if every service in the land who wanted to make bot readable info available produced it in the same XML format. ( perhaps with matching csv output .. the data after the 5th comma is always runner name for example. )

-punterprofits

Have a look at microformats, see if there is anything in there which you might be able to adapt - similar sort of idea I think.

As for the football Inksmithy  .. I also happen to run a tip site with bloke who as his day job works as an odds compiler for a major uk bookmaker. Issues with that at present may be with some of his bet types not being 100% suitable for exchanges. ( some pretty illiquid markets further down the leagues / the odd asian handicap bet at a bookie perhaps with no direct tie up to betfair asian quotes etc  )
Obvious possibilities however to develop there a more exchange only bot friendly type service.
All the best
Mick

-punterprofits

I'm not really looking for tips, I'm looking for scores which are updated every ten to 20 seconds or so - I'm not trying to beat betfair, I'm more looking for another metric to include in my calculations. I've tried some webscraping, but found I had to keep using different proxies because they would block me, which is irritating. I might try that again though, with a better set up timer....thats an idea...hmm.

Use of live scores is an interesting angle Inksmithy

if you can bet successfully in play you have much more opportunity to turn over more volume.
Arguably as well their may be more opportunity to find value.
eg take the prem league 1x2 markets.
With markets up for most of a week and lots of bookmaker statisticians compiling odds
it may be hard to find value.

The amount of knowledgeable in play money may diminish however after the off.

wonder how the bookmakers who offer in play betting do it.

Most likely they have an automated in play odds engine that may take in inputs such as pre off 1x2 odds, time elapsed and current score.

I can see you perhaps working in for arguments sake 5 min blocks where you recalculate odds, post them up at your desired prices .. remove them again if not matched in your agreeable time frame.

Obviously having the correct current score is key.
Even 20 seconds can be a long time live in play.
As Hanson would say .. "It only takes a second to score a goal"   :)

quick brainstorm for you that you may already have considered..

could you set up multiple pulls of live score data from different locations.
eg picture 5 or 6 locations from which you pull at a rate the original site is content with.
Your main program then pulls not from the orig source but from your own locations and takes as current the one with the freshest timestamp. ( to cover against different server times .. the score sum with the highest total )
orig host is happy as they see 5 or 6 different pull urls at an agreeable rate.
you then suck in all of these and merge to get your end result.. info at a rate you need.

I am not super technical as you can probably tell.
far from sure if that is a good approach or not.
if of any use to you however I have a few urls on different hosts and would be happy enough to upload a simple little scraping script for you as part of your chain. see site link above and use admin@ email.

Another concept may be to remove any necessity for outside score info all together and create some form of betfair data only score prediction routine based on perhaps market in play suspensions and sudden price shifts.
No this would not get it right 100% of the time  . but nor would a method dependant on possibly eronious third party live score data.

All the best
Mick

Mick - the odds on the correct score market could give you some concrete information on what the current score is.  I think the odds on the current score being the final score will shorten during play.  So poll that market and see which result is shortening and that should be your current score.

Well that was a load of rubbish.  Several of the scores shorten.  Looking at one match this evening, the 0-0 and 0-1 both shortened from 12 to 5 during the first half without a goal.  In fact the 0-1 shortened marginally more than the 0-0. 

Can you not work out the score from which selections on Correct Score have ridiculously high odds?
For example, if the score is 1-1, you'd expect 0-0, 1-0 and 0-1 to be backable at 1000, while 1-1 will have normal odds.
In general, if ridiculously high back odds are available for m-n, then EITHER the home team has already scored more than m, OR the away team has already scored more than n
Conversely, if lay odds are still available on score m-n, indicating that someone is willing to back it, you know it's still an achievable score i.e. the home team has score m or less goals and the away team has scored n or less goals.
Putting these two together you should always be able to work out the current score?

could you set up multiple pulls of live score data from different locations.
eg picture 5 or 6 locations from which you pull at a rate the original site is content with.
Yourmain program then pulls not from the orig source but from your ownlocations and takes as current the one with the freshest timestamp. (to cover against different server times .. the score sum with thehighest total )
orig host is happy as they see 5 or 6 different pull urls at an agreeable rate.
you then suck in all of these and merge to get your end result.. info at a rate you need.

-punterprofit

I have actually considered doing that and I think in all likelyhood its what I will end up doing, so long as I can find a few sites which get their feeds from different places. I haven't done it yet because I've been busy with other things.

Looking at the correct score market is a useful metric as well, since it not only gives you the opportunity to deduce or infer the current score, it also gives you the ability to see which way the rest of the crowd thinks its going to go. I wouldn't consider relying on it though. What it could be used for though is to come up with a metric whereby all of these different factors are used to come up with a tipping point. Give all the metrics you are using a 'weight' which depends on the reliability and recency of the data you are using and when the combined 'weight' of all the metrics (IE, weight of money, volume traded, how quickly the market is changing) reaches a certain value, the betting process is started. Whether that involves placing a back bet then placing a lay bet a few ticks later is up to you.

Lots of factors can be involved then and could probably be organised so new metrics can be added reasonably easily.

Damn, so many ideas, so little time!

INKY
you can get a free feed from http://www.livescore.in/free-livescore/
set up some sort of leech from that feed and they wont block your IP .

cheers deano, I'll have a look at that - I got up to the point where I'm looking at the rules and regs.