Hi
Test Mode is brilliant for testing the infrastructure of your trigger and it simulates the exchange quite well, but of course is not the exchange. The key differences as described in the help file are:
- test bets cannot be split or partially matched
- your test bet will always be matched if the price is available in the market, irrespective of the amount of money available at that price
ideally one would develop and test a trigger using test mode, and then use small stakes to do further testing in real mode.
and just a few other pointers, especially for in-play betting where prices can move in a millisecond :
- in real mode there is a latency for your bet getting to Befair, internet connection etc. there is a Test Mode setting you can adjust to emulate a small delay:
delay placing and cancelling test bets by x seconds. I have mine set to 0.8secs. - some markets have an in-play betting delay imposed by Betfair
- if you bet below Betfair minimum the bet will take approx. 3x longer to get to Betfair than a bet at or above Betfair minimum amount.
Especially in-play, the price you trigger at may not be available by the time your bet gets to Betfair, consider offering prices several ticks higher or lower than the current price to give it a better chance of getting matched, or using "best" which Betfair will match at the best available price. This is the helpfile description of "best":
The term "best price" here refers to the lowest possible price available for your whole bet to be matched. Let's say you lay £1000 at 4.5, but only £500 is available at that price. Another £400 are offered at 4.4 and £300 are offered at 4.3. If the program is to match your whole bet, it will have to lay £500 at 4.5, £400 at 4.4 and £100 at 4.3. Luckily this is done automatically by the exchange if you lay at a price that is higher than the current available odds. I.e. if the program offers £1000 at 5.0, the exchange will automatically lower down that offer to the best available prices and match all three portions.