Time variables

Hi,

My question is over the best and most accurate way to calculate the time since the off.

I know at least 5 different ways to do it.

1-match_minute
2-market´s minutes since the off
3-(now_time-event_off)*mf_minute
4-(now_time-last_inplay)*mf_minute..(horses, greyhounds...)
5-(now_time-OFFTIME)*mf_minute, being (OFFTime=now_time) an user variable set when the market turns inplay


I noticed that match_minute isn´t identical to (now_time-event_off)*mf_minute (even in the 1st part of a football match). Also that the updating of match_minutes is, apparently, not connected with the refresh rate. It´s not constant and only a few times per minute independently of the refresh rate.

An extra question would be how to calculate accuratly the time of the 2nd part in a football match.

Summarizing: what´s the best way to determine both the total time of an event and, in the specific case of football, the time of the 2nd part.

Thanks

Hello!

I noticed that match_minute isn´t identical to (now_time-event_off)*mf_minute (even in the 1st part of a football match)

That's right. Firstly, the event may turn In-Play, but the minute of the match is determined by referees/venue owners.
Also, at the end of the first half, the minute of the match no longer reflects the time that has passed since the beginning of the game.

It´s not constant and only a few times per minute independently of the refresh rate.

Like all score data, the match minute is updated by third-party operators. MF Pro checks for score and match minute (across all football markets loaded into the program) once per 5 seconds, regardless of the market refresh rate.

The answer to your question is: use the match_minute variable at all times for football markets. Use match_extra_time for any extra time at the end of the current half.

If it is not about football, then use (now_time-event_off)/mf_minute.

Hi Oxa,


I have some doubts about your answer

That's right. Firstly, the event may turn In-Play, but the minute of the match is determined by referees/venue owners.

I agree, but after the referee has started the match, the clock switchs on, and, supposedly, it doesn´t stop till the end of the half. Then the referee decides the necessary injury time. So, at least in theory, the result should be the same regardless of the variable you choose.

When you say that match_minute is more accurate you are assuming that event_off, or an user variable, set to when the market starts, may not be set exactly at the off (the market may not turn inplay exactly when the referee starts the match. Am i right?

My problem with match_minute is that, even if the MF PRo, updates it each 5 seconds, in pratrice, the change in the values is much slower probably due to the data itself. In the tests i made the update rate, even in the same match, wasn´t constant, and could go often up to 1 minute. In Time sensible situations like extra-time, a more accurate measure may be fundamental.

If it is not about football, then use (now_time-event_off)*mf_minute.

I was thinking to use (now_time-last_inplay)*mf_minutes in horse and greyhound races to avoid problems with false starts. Which one do you recommend?

Thanks,

I made a typo in the formula, should be:

(now_time-event_off)/mf_minute

I've corrected my previous post to avoid confusion. Yes, this formula is the way to go with races.

In my experience with basketball, for instance, match_minute is as good as useless: sometimes it won't change for minutes.

With football it's much better though: I've never heard of it not changing for a whole minute. Can you share the results of your tests, i.e. in what matches was this variable inaccurate?

The discrepancy between the time that has passed since the off and match_minute is very insignificant: seconds, half a minute max. But that's until the end of the first half.

Hi Oxa,

I made a typo in the formula, should be:

(now_time-event_off)/mf_minute

You are right. My bad!!! I would correct in my post but i don´t know how to do that.

The question i place is: and if there is a false start? Will the event_off restart with the new value? That´s why i was thinking that "last_inplay" would be better.

With football it's much better though: I've never heard of it not changing for a whole minute. Can you share the results of your tests, i.e. in what matches was this variable inaccurate?

Till now, in no match the rate was constant or, at least, close to 5 sec. To give you a better information, i made some random tests with the current inplay matches
 
Barcelona v Leganes - update from 64.38 to 65.18
Sint Truidense (res) v Anderlech (res)-from 24.03 to 24.72
Colchester U23 v Reading U23- from 26.77 to 27.55
hamburger SV v Nurberg - from 4.5 to 5.18
Breidablik v IA Akranes - 80.95 to 81.50
KR Reykjavik v Vikingur Reykjavik - from 40.83 to 41.22
Birkirkara v Banik Ostrava - from 35.87 to 36.10
Europa Point FC v Glacis Utd - from 10.85 to 11.23.

The discrepancy between the time that has passed since the off and match_minute is very insignificant: seconds, half a minute max. But that's until the end of the first half.

Yes, you are right the difference is, at the moment of the update of match_minute, usually  less than 30 seconds. But match_minute seems to be updated, i would say in average each 30 seconds, and worse than that in a not constant rate. So, the difference can be greater than 1 minute and not fully predictable. For the strategy i want to implement that´s not enough.

I will make some tests to see which one is more accurate and, if it´s match_minute at the moment of the update, i will probably go with an hybrid solution. An user variable based on an one shot of match_minute but updating at the second. I still need to materialise it.

Thanks,