Knowing and monitoring the start of betting?

In sports  particularly, it's impossible to know when trading opens? Very dependent on the popularity. For a regular football match, I guess it's a week out. For an insignificant tennis match, it might be a few hours before the start. 

1/ I want MF to monitor exactly when selected markets Open, and then start responding to prices from the get go. Not knowing when a market opens can put you at a disadvantage if you are thinking of using a system that relies on the markets response to the initial odds offered on two teams.

2/ Does Betfair open the bid with some sort of notional set of prices? There's always a spike at the beginning of price history graphs as the market participants adjust for their take on the "correct" pricing. If Betfair don't kick things off, then a punter must, but he/she/bot needs to know exactly when the market opens?

3/ I suspect that no one knows the answer, as I've begged the question on many forums and to Betfair directly (for no reply). For MF to know when the first trade is placed, it has to know about the match even occuring? So when (time scale) is the schedule of matches released for all the world to start betting?

hi
you can set mfp market locator to look for events 1 to however many days you want to look , the locator will then give x number of markets to load.
you can also set mfp to load markets every however many hrs or time or/and on opening mfp.

mcbee

OK mcbee, that's interesting. Thankyou. I haven't used MF for a few years, but might revisit? So if backtesting a strategy on past data, is it possible to right a rule like:
The second the first trade is made, enter the market with bids based on an over round of x% or some other condition? It's capturing the moment of the first trade, that's the crux for the rest of the system.
Cheers

hi
yes but you would have to monitor the markets from day 1 at a slow refresh rate because of the days it will be monitoring and the amount of markets.
you might find it better to use maybe 4 profiles running different markets.


mcbee

In the attachment, you can see that it's the very early price action of this soccer match that's where it's all happening? As you'd appreciate, without the benefit of hindsight, you don't know which way it's going to break....lol...but that's the enigma we face every day? Double whipsaws, you name it, can and do happen. Not for the fainthearted.

You're right about needing lot's of "instances' running concurrently. That's where software and internet download speed can become a real handicap, in just coping. Last time I used MF, I gave up on it's integration with excel as it was too unstable. Other software coped much better. Not sure if anything's changed over the last couple of years with MF's development?

Probably a bespoke program using Python or suchlike could handle it better, but that's a mountain to climb in itself, if like me you're not that savvy with writing code.

Thanks again for your input.

Hi
As mcbee suggests you can monitor the markets as soon as as they are available using the market locator.

capturing the moment of the market's first trade is:
markets matched volume is greater than 0
though I think it would be better if you adjusted this to your requirement and to avoid minor fluctuation.
for example, looking at your graph and the laying of the home team early on, I would suggest:
markets matched volume is greater than 100
before trading

Thanks MarkV. I will pursue it. Interesting no one has commented on my inference about, "Who are these lurkers that place the first bet?". Is it something they specialise in, or is it just a hack at Betfair wanting to kick things off? Whoever it is, they are obviously testing the water with their initial bids. Further, if the trigger is a volume of >0 then you will never be the first trade. How could you start things off with MF? In other words place the first bids, you need to know when the market is open? You also need to have done a rough odds assessment. When Betfair open a market for the first time, do they kick it off with a bid themselves and our bots simply respond? Interested to hear other peoples views.

Hi
I don't think Betfair seed the Sports Exchange markets. I think the first traders in a market would be looking for a value bet to get the best price they can.

In the attached screenshot I do a market locator search for MO markets with 0.00 traded volume, it finds 170 markets.
In the market I look at there is no trade yet, but you can see the away team is the favourite as the market is forming.
In this market I would offer a back bet on Southern District somewhere in the region of 1.78 to 1.85 say?
as the prices get more competitive, someone will match what is on offer, and there's the first trade.

further to the above, if you wanted to continually offer the best price in the market until you are matched, you could use a trigger like this one:
http://marketfeederpro.com/learn/triggers/best-opposite-price/

That's very interesting MarkV and thankyou for your time in following this through.
I like your trigger suggestion but would modify it, such that we're bidding on all three selections concurrently. That is three Lay bids for a 101% book and three Back bids for a 99% book. We are the market maker after all!...lol The bot will continually cancel and resubmit until a first match is achieved.

Taking the game above I used as an example:
The trigger file assesses the current bids and proportions them as 43.44:26.30:30.26 = 100%. Then creates overs/unders and submits:
To Back  $2.32:$3.85:$3.33
To Lay  $2.27:$3.77:$3.28

If a Back is matched first then all lay bids are cancelled and you run with the remaining back asks and vice versa. Looking at those charts you should be able to green the book fairly quickly. Now the bot sits back and this next part is the most difficult trading decision? At what level do you take profits? With the benefit of hindsight you could optimally squeeze maybe 9% out of this market? It's a trial and error game, but you're not sitting there watching this all unfold. Ideally you should have 70 odd of these games running concurrently, which brings me to the next question? Is that possible with MF??

Hi
Don't see any problem with 70 or so markets being monitored at a slow refresh rate as mcbee suggested earlier. I similarly run triggers on a good number of concurrent OU markets at an idle refresh rate of 45 secs and it works a treat.