Using Betfair historical data

[Trader_Tom]

Can anyone tell me how I can make use of the Betfair historical data? I've downloaded the csv file but I've not idea how to view or make use of the data!

All the best,

Tom

[workpeter]

Well you need to organise the data first. this is simple.

1) open the document in excel. Chances are all the data will be in one column. Your objective is split the data across multiple columns.
Example column A) Horse Name B) Starting price c) results

To split data goto data > text to columns > delimitted > you probably would want to select comma. Most CSV data files split their data using commas.

2) write a basic betting algorithm

In our basic example columns A,B, and C is your historical data. So use column D to write a betting formula. I have attached a screenshot and basic spreadsheet with a demo.

the basic concept is to define a betting formula to be applied across the entire data range and hopefully have a profit at the end.

[LinusP]

Have a look at the SP data betfair provide if you are interested in racing, much easier to work with:

viewtopic.php?f=5&t=11883&p=100709#p100709

[Trader_Tom]

Well you need to organise the data first. this is simple.

1) open the document in excel. Chances are all the data will be in one column. Your objective is split the data across multiple columns.
Example column A) Horse Name B) Starting price c) results

To split data goto data > text to columns > delimitted > you probably would want to select comma. Most CSV data files split their data using commas.

2) write a basic betting algorithm

In our basic example columns A,B, and C is your historical data. So use column D to write a betting formula. I have attached a screenshot and basic spreadsheet with a demo.

the basic concept is to define a betting formula to be applied across the entire data range and hopefully have a profit at the end.

Cheers Peter! I'll have a look at it later today : )

[Trader_Tom]

Well you need to organise the data first. this is simple.

1) open the document in excel. Chances are all the data will be in one column. Your objective is split the data across multiple columns.
Example column A) Horse Name B) Starting price c) results

To split data goto data > text to columns > delimitted > you probably would want to select comma. Most CSV data files split their data using commas.

2) write a basic betting algorithm

In our basic example columns A,B, and C is your historical data. So use column D to write a betting formula. I have attached a screenshot and basic spreadsheet with a demo.

the basic concept is to define a betting formula to be applied across the entire data range and hopefully have a profit at the end.

Cheers Peter! I'll have a look at it later today : )

[Trader_Tom]

Have a look at the SP data betfair provide if you are interested in racing, much easier to work with:

viewtopic.php?f=5&t=11883&p=100709#p100709

Cheers : )

[Atho55]

Tom

IMHO you need to decide what you want to get out of it and how much time you want to devote to the compilation and the continual updating. Having just compiled roughly 2 years worth (from another source) I consider mine to still be "work in progress". However, once you have it completed you can sort (Excel / Pivot Tables) to give you pointers towards the days trading.

Glyn

[nigelk]

Using the timestamped Betfair data can be a bit of a nightmare due just to the sheer amount of data they contain. Just pulled out a random sheet from 13th Jan 2014 to the 19th 2014 and it's got 592,584 rows of data.

However, if you decide exactly what info you need you can cut the size down by a huge amount.

I've removed all non UK/Irish races, got rid of any special bets, (distance between 1st & 2nd etc etc) and just concentrated on the odds movements in last 10 mins before the off.

To make the row size even smaller, the odds have been transposed from rows to columns, so instead of 1 horse taking up 10 rows, it now only uses one. The upshot being that it is now only 224kb and only contains 1614 rows of data.So working on the assumption of 2,000 rows of data per week, that's 104,000 rows in a year, still smaller than 1 weeks worth of original data!