Monday 18 August 2014

What to do about Gowi


I concluded last month that the next post (this one) would be dominated by two services, Club Gowi and Tennisratings, but I feel I am ready to pass judgement on the former, but not quite the latter. Having now collected 6 months worth of bets from Gowi, it felt like the perfect time to delve deeper into the numbers.

First I thought I would run Gowi through my 8 point checklist. When I wrote this list I was anticipating Gowi falling at a few hurdles, but as I mentioned they are not the be-all-and-end-all, just starting points.

1.              Clear selections with relatively stable delivery times – most emails arrive at the designated time and Keith is very good at keeping the customer aware of the schedule ahead.
2.              Bets and prices are achievable at my limited list of bookies – In terms of being available I would say 90% of the bets are at Asian style bookies, most of my bets are with Pinnacle or Matchbook. The ones that are not are the issue as they are usually the bigger priced outright bets, and I have really struggled to get any significant amount of money on these selections. Availabilty is also a big issue, but that will be addressed in the rest of this post.
3.              Reasoning is clear even if result of algorithms/rules – there is no doubting Keith’s reasoning for the bets, if anything there is too much text, but he obviously works hard and there are very few bets that give the impression of being rushed. The only one that springs to mind was “Limerick "always" beat Athlone, regardless of the situation, they just do !”, before promptly losing 3-0!
4.              Service is stable in terms of psychology and staking (not on tilt) – I would say this is true although I do feel the bigger priced golf and outright selections are over staked, especially in relation to the bets I have received, but I suppose I have not seen any of them win!
5.              Records are clear, verified and analysable – Afraid not, and anyone who follows SBC or Daily 25 will not be in the least surprised by this decision. Yes there are graphs and summary numbers but nothing that one can get one’s teeth into. I know from emails he has sent out that Keith does not think it is relevant, but I would consider my view to be very different to this. In fact I hope that my post helps other people to get a good feel of the service, for better or worse.
6.              Profitable in the long and medium term – All the numbers point to this being so, but there is always the nagging doubt that price collapse make it a challenging task, like so many other successful tipsters, if only we had the data to show it …
7.              Minimal clashes with others in stable – This has been a bit of an issue for me. I have had great success with big priced selections in Europe, and a few times Keith has been on the other side giving away goals on the handicap. Also WTA tennis has seen a bit of a differing of opinion with Tennisrating, but it is certainly not at deal-breaker proportions.
8.              Price is justifiable – Now that Skeeve has taken a back-seat, this is by far the most expensive service I follow. I think it can be justified by the amount of hours that is put into the previews, but it might just be me, but I always hold the more expensive ones up to a much stronger light. In a way, they have to prove themselves over and over again, especially if I can’t analyse the numbers and create a picture for myself.
 
So: 8 checks, 2 ticks and 1 cross, 5 not sures. To be quite honest before I finished all the analysis I was fairly sure I wouldn’t have the ammo to beat those sort of marks, and Gowi would be going by the wayside, but lets give it a fair chance.

Before I give you the data, I would like to start by saying I have made a slight loss so far following Gowi. This, as you will soon see, is not of Keith’s making as I do not follow religiously and have struggled to get certain bets on. There have also been periods – such as holidays – that have made it very hard to place bets, so I have often taken a break.

To give a truer impression of the service I decided to go back and record all the tips, and where possible, collect the closing prices from Oddsportal. Now as you can imagine this was quite testing so I decided only to collect the football prices that were available at Pinnacle. That means BTTS and team Over/unders selections are just recorded as advised price. I could have done tennis as well but was losing the will to live at this point so, as football makes up by far the biggest proportion, this is what you get!

I also decided to leave out the few bets that are advised in-play, such as

I do expect Brazil to win and maybe everything will click today, but we have no real need to bet, as match odds do look a little skinny, however, should Brazil trade at 2.20 "in running" , whilst still 0-0 and as long as we are 11 v 11, I will take that for 1.25 units.

I do remember thinking when Keith decided to pull his proofing from SBC that these types of bets would be a controversial area. From a proofing point, it is not only hard to decide if the price was achieved, it is also open to manipulation as to whether it was achieved. If the price touched 2.1 and it won, would it be included and if it lost, not quite, that sort of thing! I am not accusing Gowi of any of this, it is just without the data and a list of bets, these thoughts do cross your mind.

The best place to start is an overall picture, so in the 6 months from the 22nd January to 21st July there have been:
                                               
                        Bets                             Stakes                         Adv profit                  ROI

659
848.23
24.8
3%

If we just look at the bets I have used closing prices for, ie football win, AH and O/U we get:

                        Bets                             Stakes                         Closing profit              ROI

449
618.83
12.87
2%

Interestingly the profit on those equivalent bets at advised price is 22.77 points or 4% ROI.

Before we get on to the bulk of the football bets, lets look at the two other sports that I have seen during my subscription (there has been a very successful American Football ante post success that I noted in this time period, but I was not part of as it was placed before my time). Firstly golf has seen 78 bets placed, for 81.5 points and a loss of 8.82 points.

Type
Bets
Stakes
Profit
ROI
Pre-start
24
26.25
-15.075
-57%
In-play
41
40.5
1.3375
3%
Match
13
14.75
4.9175
33%

As you can see from the table above, the most successful area has been matchbets (including 2- and 3-balls), while pre-tournament advice has yet to hit the mark. Saying that, at high odds it will only take one to put that area into profit. I do have a feeling that the staking is too heavy on these bets, with an average of over a point a bet, and average odds of over 14/1, including top 5s, 10s and 20s.

Tennis has produced 57 bets, 67.25 points staked and have turned a profit of 4.4 points at an ROI of 7%. I must add from my point of view that I am still in negative territory here as Petra Kvitova’s Wimbledon win provided an overall profit of 5 points but I could not place the bet at the recommended price with my bookies.

Type
Bets
Stakes
Profit
ROI
Result
15
18.75
7.41
40%
Correct score
18
13.75
-1.81
-13%
Handicap
17
24
-1.50
-6%
Outright
3
5.25
3.50
67%
O/U
4
5.5
-3.18
-58%

The strongest area has been outright winners (by ROI) and match winners by profit. So far, any attempt to improve the odds by predicting the correct score, game or set handicaps or over/unders have a negative ROI, but the sample is still very small. Outrights would have been boosted further as Na Li won the Australian Open but alas I was not a subscriber when the tip was issued.

So golf and tennis have staked about 150 points between them leaving just over 700 points for football, which at advised prices returned a profit of 29 points at 4% ROI.

To be honest, that is not a bad return at all, especially give the hard time he is having on forums and blogs recently but I suppose the question of whether it is achievable is key here. My record of comparing closing price to advised price has shown a near halving of profit level and ROI. There is also the issue of the price movement as soon as the bet is released, which I would claim demonstrates a bigger gap between advised price and the odds, say 1 minute later, than the previously mentioned one.

In defence if Keith, he has always maintained that if you do not get the price at issue, aim for nearer the off or play in-play, and that certainly seems to ring true with my findings.

Row Labels
Count of Movement
Sum of Stake
Advised profit
Closing price profit
-10--15%
33
43.3
8.5
3.4
-15% or worse
14
16.3
11.0
5.8
-5%--10%
93
131.8
-7.4
-15.0
>10%
31
43.3
0.1
5.9
0--5%
126
175.5
18.2
13.6
0-5%
107
148.8
-0.7
2.8
5-10%
45
60
-6.9
-3.5
Grand Total
449
618.8
22.8
12.9

Those prices that have jumped off a cliff have still been profitable (over 10% reduction) with over 9 points profit, although those few who got the price achieved nearly 20!

The area of concern is when the price fell by between 5% and 10%, although these bets were losing anyway. I must add these are pretty small samples so not to be taken as a rule.

Bets that go off just 5% worse or better (a price increase) are in the closing price bettors favour, namely 18 points of profit against 10 points at advised. Quite clearly this is an improvement on the total return at starting price (18.7 vs 12.9) and you will still be betting on nearly 70% of the advised selections.

This pretty much fits with what Rowan has said in his Practical Punter column – 5% max drop on negative handicaps and 10% on positive, and Steve at Daily 25, who plays a maximum drop of 7%.

The next angle I looked at was type of football bets, slightly inspired by Rowan’s positive/negative split. For sake of argument I included AH -0.25 and up as negative and level ball or DNB or better as positive.

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
221
307.5
6.9
303
0.5
AH+
145
196.3
4.6
195.5
1.4
BTTS
11
15
0.5
0

draw
5
4.1
-4.1
4.1
-4.1
goalscorer
9
9.5
17.0
0

HT
3
3.3
-3.3
3.3
-3.3
HT/FT
8
4
0.3
0

O/U
85
116.5
22.7
113
18.4
outright
24
30.9
-18.2
0

Teams O/U
13
12.5
2.7
0

Grand Total
524
699.48
29.19625
618.83
12.8725

Well from this I can see the biggest area of profit seem to come from goal betting, both overs (as most are overs) and goalscorer bets. The goalscorer bets include both anytime match bets and a series of top goalscorer bets in the World Cup. Asian handicap bets (including straight wins) are profitable at advised prices but less so at closing price.

Looking further at just the previously mentioned split of -5% or better we get:

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
155
218.3
-6.4
218.3
-2.5
AH+
97
131.3
7.6
131.3
10.8
draw
3
3
-3
3
-3
HT
2
2.5
-2.5
2.5
-2.5
O/U
52
72.5
14.9
72.5
15.9
Grand Total
309
427.5
10.6
427.5
18.7

We might as well see the counter example, where the price has fallen by over 5%:

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
62
84.8
12.6
84.8
2.9
AH+
47
64.3
-4.9
64.3
-9.4
draw
2
1.1
-1.1
1.1
-1.1
HT
1
0.8
-0.8
0.8
-0.8
O/U
28
40.5
6.3
40.5
2.5
Grand Total
140
191.3
12.1
191.3
-5.9

Looking at these tables, maybe Rowan is wrong in extending the price drop with positive handicaps to over 5% and should only do that with negative ones!

Seriously though, at -5% or better most areas are positive, and negative handicaps and over unders still produce the goods at worse prices. There may even be an argument for taking a less favourable line in these cases, but I haven’t fully checked that!

After value, I looked at the different competitions that Gowi bets on. Out of all the tipsters, Keith has by far the biggest spread of leagues to cover so surely some might be better than others.

I mentioned in my checklist that there were a few clashes with other services, so my first comparison point was with those.

The bulk of the system bets come in the UK leagues, and methodologically the biggest clashes seem to be when rational argument meets data, so how does Gowi do in the UK.

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
58
76.3
-5.8
71.8
-6.7
AH+
35
47.0
4.6
47.0
4.3
BTTS
1
1.5
1.5
0.0

draw
1
1.0
-1.0
1.0
-1.0
goalscorer
1
0.5
-0.5
0.0

HT/FT
3
1.5
-1.5
0.0

O/U
13
17.5
3.4
17.5
3.0
outright
1
1.8
-1.8
0.0

Teams O/U
6
5.5
-1.0
0.0

Grand Total
119
152.5
-2.0
137.3
-0.3

Well is that worth the stress of placing the clashing bets? Especially as they always arrive at least the day after TFA and FI. It looks like the AH+ bets are solid, and these are less likely to oppose the systems, which on the whole favour the outsider, and overs are profitable, which don’t clash, so there are some angles to consider.

The next area of clashing is in the Top flight European leagues, in this case with FI’s Top Flight Euro selections and also Football Elite.

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
52
78.8
9.7
78.8
7.0
AH+
25
35.5
-3.4
35.5
-4.0
BTTS
3
4.3
-0.9
0

draw
1
0.3
-0.3
0.3
-0.3
HT/FT
2
0.5
-0.5
0

O/U
22
31.5
4.2
30.3
3.2
Teams O/U
2
1.8
3.3
0

Grand Total
107
152.6
12.0
144.8
5.9

That looks better, although all the profit is earned in France, an area of strength for Gowi as he tips in Ligue 2 as well.

Here is both French leagues:

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
46
66.3
16.1
66.3
14.3
AH+
16
23
-5.4
23
-5.9
BTTS
2
3
0.4
0

HT/FT
3
1
-1
0

O/U
30
43
12.6
41.8
9.3
outright
2
3.3
-3.3
0

Grand Total
99
139.5
19.4
131
17.6

Back in February’s monthly review I commented about the service:

“I have been particularly impressed with Keith’s tipping in domestic cup matches, which he seems to play with a very strong ‘motivation’ factor to the fore.”

So how does this manifest itself over 6 months. In domestic cup competitions from all over the world:

Row Labels
Bets
Ad Stake
Ad Profit
CP Stake
CP Profit
AH-
25
35.5
1.7
35.5
1.9
AH+
33
43.3
11.8
43.25
11.5
BTTS
3
4.3
-1.3
0

goalscorer
1
0.8
1.5
0

O/U
3
3.5
1.8
3.5
1.4
outright
3
4.9
0.7
0

Teams O/U
4
4
-1.4
0

Grand Total
72
96.2
14.9
82.3
14.8

Wow, I think I was onto something! Judging by the AH+ returns, Keith does very well in picking the best way of playing the underdog. As this is highly unlikely to clash with any other service, it seems an absolute no brainer to play these.

There are other angles that I have looked at but I am probably boring you all now. Since doing all the research, my subscription run out and I took the decision to rejoin and Keith kindly upgraded a 6 month to an 8 month subscription, so that should take me to the end of the domestic season, or as near as damn it.

I still think I will play as much as I can but will certainly look to get value and not dive in at posting time. I find it quite refreshing not to worry about posting time and no longer wait by the computer at 10am most mornings. Now that I believe that they will either rebound or not represent much value, I can wait for a later entry point.

My focus will be on all types of goal betting and cup competitions but until I have a much greater sample, nothing will be ignored unless I can not get a fair price.

What has been a struggle has been ante-post bets on this season’s football. My limited set of accounts do not cater for very big bets at fancy prices so I have been unable to get more than £10 or £20 out of £150/£175. Judging by the last 6 months this shouldn’t be a big issue, but I do get the feeling they work out well in the long run. Certainly subscribers that were around in the few weeks before I started are sitting on a windfall thanks to Rory McIlroy. In fact that makes three near jackpots missed with Seattle Seahawks and Na Li thrown in in January!

Next post

Finally getting a grip on Tennisrating, and a good job too as his tips service will stop later this week. I have found a few angles to follow and some not to, so nearly ready to discuss, although one angle is so good I am tempting to keep it to myself as it is so good!

3 comments:

  1. Good blog but I feel compelled to give some balance

    Opening paragraph: you say ‘I feel I am ready to pass judgement’…surely you can’t pass judgement on any service with just 6 months’ worth of data. It’s simply not long enough and I believe you need of minimum two seasons.

    Point 2: why do you only have two bookies – how come you don’t have more Asian bookies, do you not use a bet brokerage service such as Sportmarket or Asian Connect.

    In terms of the outright selections, I know full well how difficult it is to get on with traditional UK books but surely you could have some sort of plan in place ahead of the release of the bets…something like using a friend’s account or being near a bunch of different betting shops in a town. I think most people would be able to put some sort of plan in place whereby they get some of the prices. I know I got on all the bets by having access to a number of different books and having oddschecker open.

    Point 3: Not sure what you’re getting at with this point ‘reasoning is clear even if result of algorithms/rules’.

    With reference to some bets being rushed….your example was the Athlone v Limerick. Was this bet really rushed? He gave two paragraphs where he mentioned head-to-head and recent form. Also, the double inverted commas around “always”….come on man, are you seriously using this as a credible example?

    Point 4: I don’t see how you can feel that bigger priced golf and outright selections are over staked…are they out of line with his daily football? This also brings me to a topic about people feeling the need to bet everything that Keith’s suggests. Surely you should be able to work out what’s best for you in his service – Golf, Tennis, NFL etc. It seems to me that you may be struggling to deal with the variance of his outright golf selections.

    Point 5: Ok, so this point has been discussed in great detail and here’s my take on it.

    Keith has stated on a number of occasions this his service is not a tipping sheet and I feel many are struggling to accept this as they automatically assume he has to be categorised as a tipster and therefore has to publish his results, has to get his results verified, has to do this, has to that etc etc.

    Why is it not acceptable to say ‘I’m more than a tipster. I’m a newsletter service who provides detailed insight into football matches and sporting events all over the world. I give recommended bet suggestions on these sporting events’

    How many of Keith’s ‘new’ subscribers just simply scroll down to the bottom of each newsletter to find the red type bet? I reckon about 90% of his new followers don’t even read the actual content and miss out on numerous other betting angles…but hey they don’t get this because they’re too busy chasing prices.

    The fact that Keith has been providing a winning service since 2006 suggests that he’s slightly successful in this game!

    Point 6: Price collapses…there is always going to be a price collapse if everyone is trying to get the units down at 10am in the morning - two key alternatives is to wait near the off or go in-play. Keith has mentioned this numerous times.

    ‘If only we had the data’ – if you choose to follow the service you will eventually have data

    Point 7: feel this is irrelevant as if you choose to follow a certain ‘tipster’ then you should do so, regardless of conflicting bets, and just do your analysis at the end of the season.

    Point 8: I’d like to see Keith increase the price of his service and restrict the number of subscribers!

    ReplyDelete
    Replies
    1. Hi Franck, thanks for the post, just written a long reply but have been told it is too long to post. Going out now so will either edit it or turn it into a response post on the blog.

      Delete
  2. Hi ,
    Great work going through all the stats. Do you have the stats for all the big leagues PL/La liga/Serie a/Ligue a ? I do not have enough stats to make anything out of it but the little I

    have suggest that these are losing leagues all together. And I am only talking about the AHCP , totals are in profit all over the leagues. The question Is If not the profit had been higher

    If you would only bet Japan/Brazil/Usa/Ligue 2/Domestic cups. And what about the numbers in the english leagues without PL ?
    My stats also points to that there Is now value in following the bets from the irish/scotish or any other league with only small number of bets.
    As I said I have only small samples so hopefully you have a little more. And once again only talking about AHCP.
    Cheers !

    ReplyDelete