Soccer Predictions using Python (part 3a) by stevencurrie

View this thread on: hive.blog | peakd.com | ecency.com

programming · @stevencurrie · Sep 23 '17

Soccer Predictions using Python (part 3a)

Just posting the latest output from my Soccer Prediction program.  Too late at night (early in the morning?) to write up an explanation of the changes I've made to the code but I'll get a new article posted later today, after I get a couple of hours sleep. 

Nighty night.

Match|Prediction|Probability|Odds
-|-|-|-
Everton v Bournemouth|Everton|79.64|1.26
Man City v Crystal Palace|Man City|71.04|1.41
West Ham v Spurs|Spurs|85.95|1.16
Rangers v Celtic|Celtic|75.02|1.33
Bolton v Brentford|Brentford|76.81|1.30
Derby v Birmingham|Derby|67.43|1.48
Leeds v Ipswich|Leeds|67.08|1.49
Norwich v Bristol|Norwich|78.53|1.27
Reading v Hull|Reading|81.79|1.22
Sunderland v Cardiff|Cardiff|69.19|1.45

👍 minnowpond, outcast81, stevencurrie

`author`	stevencurrie
`permlink`	soccer-predictions-using-python-part-3a
`category`	programming
`json_metadata`	{"tags":["programming","python","soccer","predictions"],"app":"steemit/0.1","format":"markdown"}
`created`	2017-09-23 02:17:12
`last_update`	2017-09-23 02:17:12
`depth`	0
`children`	5
`last_payout`	2017-09-30 02:17:12
`cashout_time`	1969-12-31 23:59:59
`total_payout_value`	0.000 HBD
`curator_payout_value`	0.000 HBD
`pending_payout_value`	0.000 HBD
`promoted`	0.000 HBD
`body_length`	691
`author_reputation`	7,132,661,654
`root_title`	"Soccer Predictions using Python (part 3a)"
`beneficiaries`	`[]`
`max_accepted_payout`	1,000,000.000 HBD
`percent_hbd`	10,000
`post_id`	15,668,732
`net_rshares`	3,300,435,550
`author_curate_reward`	""

properties (23)vote details (3)

voter	rshares	pct
minnowpond	1,217,024,240	1%
stevencurrie	911,874,355	100%
outcast81	1,171,536,955	100%

@minnowpond · Sep 23 '17

This post recieved an upvote from minnowpond. If you would like to recieve upvotes from minnowpond on all your posts, simply FOLLOW @minnowpond

properties (22)

`author`	minnowpond
`permlink`	re-soccer-predictions-using-python-part-3a-20170923t022836
`category`	programming
`json_metadata`	"{"app": "pysteem/0.5.4"}"
`created`	2017-09-23 02:28:36
`last_update`	2017-09-23 02:28:36
`depth`	1
`children`	0
`last_payout`	2017-09-30 02:28:36
`cashout_time`	1969-12-31 23:59:59
`total_payout_value`	0.000 HBD
`curator_payout_value`	0.000 HBD
`pending_payout_value`	0.000 HBD
`promoted`	0.000 HBD
`body_length`	143
`author_reputation`	13,239,048,956,578
`root_title`	"Soccer Predictions using Python (part 3a)"
`beneficiaries`	`[]`
`max_accepted_payout`	1,000,000.000 HBD
`percent_hbd`	10,000
`post_id`	15,669,353
`net_rshares`	0

@outcast81 · Sep 23 '17 (edited)

You need a really high win ratio with odds like that. I'm too dumb to understand what you are using to calculate probability, poisson distribution  you said? I look it up and it seems rocket science to me.
 You need to realise something, the odds are infact probability calculated by bookies. You are just betting your software is better than theirs. Low odds are always going to have a good win rate. If your software can sort out the worste bets in the low odd games, you are a god.
But theres something definitly wrong with your programme..Its called soccer :/

FIFA 1904
NFL 1920

: ) good luck with the bets and developing this programme, I count on you, soon we will be racing our Lamborghinnis

👍 stevencurrie

`author`	outcast81
`permlink`	re-stevencurrie-soccer-predictions-using-python-part-3a-20170923t124056996z
`category`	programming
`json_metadata`	{"tags":["programming"],"app":"steemit/0.1"}
`created`	2017-09-23 12:40:57
`last_update`	2017-09-23 12:44:36
`depth`	1
`children`	3
`last_payout`	2017-09-30 12:40:57
`cashout_time`	1969-12-31 23:59:59
`total_payout_value`	0.000 HBD
`curator_payout_value`	0.000 HBD
`pending_payout_value`	0.000 HBD
`promoted`	0.000 HBD
`body_length`	700
`author_reputation`	209,754,362,359
`root_title`	"Soccer Predictions using Python (part 3a)"
`beneficiaries`	`[]`
`max_accepted_payout`	1,000,000.000 HBD
`percent_hbd`	10,000
`post_id`	15,703,867
`net_rshares`	944,947,518
`author_curate_reward`	""

properties (23)vote details (1)

voter	weight	wgt%	rshares	pct	time
stevencurrie	0 B		944,947,518	100%

@stevencurrie · Sep 23 '17

The odds I listed are the odds my program is predicting, calculated from the probability.  Ideally, we should be able to find matches where my odds are shorter than the bookies thus indicating a "value bet".

7 correct results out of 10 seems like a reasonably good start.

Here are the results with returns for a £1 bet on each game.   28.6% profit.  If we ignored the Man City game due to the bookies odds being less than our prediction, it would have been 6/9 games correct and a 30% profit.

Match|Prediction|Bookies Odds|Return
-|-|-|-
Everton v Bournemouth|Home|1.8|£1.80
Man City v Crystal Palace|Home|1.15|£1.15
West Ham v Spurs|Away|1.65|£1.65
Rangers v Celtic|Away|1.59|£1.59
Bolton v Brentford|Away|2.16|£2.16
Derby v Birmingham|Home|1.82|£0
Leeds v Ipswich|Home|1.69|£1.69
Norwich v Bristol|Home|2.25|£0
Reading v Hull|Home|2.64|£0
Sunderland v Cardiff|Away|2.82|£2.82
-|-|-|£12.86

I'll the updated code and an explanation posted soon.

(I'm leaning more toward a Tesla myself)

properties (22)

`author`	stevencurrie
`permlink`	re-outcast81-re-stevencurrie-soccer-predictions-using-python-part-3a-20170923t192118592z
`category`	programming
`json_metadata`	{"tags":["programming"],"app":"steemit/0.1"}
`created`	2017-09-23 19:21:18
`last_update`	2017-09-23 19:21:18
`depth`	2
`children`	2
`last_payout`	2017-09-30 19:21:18
`cashout_time`	1969-12-31 23:59:59
`total_payout_value`	0.000 HBD
`curator_payout_value`	0.000 HBD
`pending_payout_value`	0.000 HBD
`promoted`	0.000 HBD
`body_length`	990
`author_reputation`	7,132,661,654
`root_title`	"Soccer Predictions using Python (part 3a)"
`beneficiaries`	`[]`
`max_accepted_payout`	1,000,000.000 HBD
`percent_hbd`	10,000
`post_id`	15,734,548
`net_rshares`	0

@outcast81 · Sep 23 '17

I see, the programme calculates odds to compare them with the ones in the bookies and find value. Thats genius ahah. For sure if you maintain this rate, you'll be rich no doubt.

👍 stevencurrie

`author`	outcast81
`permlink`	re-stevencurrie-re-outcast81-re-stevencurrie-soccer-predictions-using-python-part-3a-20170923t204939100z
`category`	programming
`json_metadata`	{"tags":["programming"],"app":"steemit/0.1"}
`created`	2017-09-23 20:49:39
`last_update`	2017-09-23 20:49:39
`depth`	3
`children`	1
`last_payout`	2017-09-30 20:49:39
`cashout_time`	1969-12-31 23:59:59
`total_payout_value`	0.000 HBD
`curator_payout_value`	0.000 HBD
`pending_payout_value`	0.000 HBD
`promoted`	0.000 HBD
`body_length`	177
`author_reputation`	209,754,362,359
`root_title`	"Soccer Predictions using Python (part 3a)"
`beneficiaries`	`[]`
`max_accepted_payout`	1,000,000.000 HBD
`percent_hbd`	10,000
`post_id`	15,739,884
`net_rshares`	935,498,043
`author_curate_reward`	""

properties (23)vote details (1)

voter	weight	wgt%	rshares	pct	time
stevencurrie	0 B		935,498,043	100%

1 reply