Saturday 9 July 2016

Shut Out

I posted a screenshot earlier this week, showing a remarkable set of results for an MLB system:

2009 : 93 - 0
2010 : 113 - 0
2012 : 92 - 0
2013 : 81 - 0
2014 : 78 - 1
2015 : 75 - 1

Almost unbelievable numbers really, and of course, that’s exactly what they are. We won't ask (yet) what happened to 2011, but the numbers are conspicuous by their absence.

As you might have guessed. the results record doesn’t relate to individual games, but to sequences of bets.

It’s a different spin on the Sports Betting Professor aka Rich Allen’s system that was discussed in this blog back in 2010.

The process is basically that the system comes up with a selection, and if the first bet loses, you chase it with ever increasing stakes until one of the following occurs:
  1. you hit a winner
  2. you run out of money
  3. you lose your nerve
  4. you reach the pre-determined maximum number of bets
Try this system with the roulette wheel some time. Wait for black to come up three times, and then bet on red (or vice-versa or use any of the ‘even’ money plays).

The chances of that first bet losing (making it four blacks in a row) are just one in 16 and it’s one in 32 that it could lose twice in a row, except that of course it isn’t.

The chances of that first bet losing are always 1 in 2 (we’ll ignore the zero to keep things simple). To recover your loss, you need to double up on bet #2, which also has a 1 in 2 chance of losing. And soon you are betting big money to make a tiny profit.

An additional problem with sports betting is that you will often not be getting even money, so if the next selection in your sequence is odds-on, you are required to more than double your stake.

The system I stumbled across is a “Shut Out” system – if a team is shut-out (fails to score) then you bet against them until they lose again, or if a team shuts-out an opponent, you back them until they win again.

In theory it all sounds wonderful; in practice, not so much.

If laying a team after they have been shut-out or backing one after shutting out another are profitable strategies, there would be no need for fancy staking methods.

Over the past 5 seasons, laying a team shut-out in its previous game would have lost you 32.27 points (-1.8% ROI), while backing a team who inflicted a shut-out in its previous game would have lost 50.37 points (-2.2% ROI). I'm not seeing an edge here.

I believe the number of bets in the sequence is set at 6, so I did a little research and found no less than six situations since last season where a team was shut-out but then won its next 6 (or more) games - first last season to do this were the Los Angeles Dodgers:
but the feat was matched by four other teams and the San Francisco Giants have already done it this season. Presumably there are other conditions to be bet, but the hazards of such a system should be apparent. 

When I started this post, I had no desire to know the exact system, but at this point, curiosity has got the better of me. The best clue is surely the one loss reported in 2014, a season when there were two cases of a team being shut-out but then winning at least six matches. In July, the Los Angeles Angels were having a great season with a record of 52-36 before being shut-out at home to the Toronto Blue Jays.
Earlier in the season, much earlier, the Milwaukee Brewers were shut-out in the third game of the season, with a 1-1 record, and then won no less than nine on the run. 
My suspicion is that there is another condition which is that the team can't have a winning record at the time of the shut-out. It makes sense - only a fool would bet against a team with a record of 52-36 - but the system creator really needs to add a 'minimum number of games' condition in my opinion.

To check this theory further, I took a look at the 2011 (missing) record, and found two 6-game winning sequences subsequent to a shut-out - and again one was the Brewers, the other the San Francisco Giants. The Brewers were 21-23 after their shut-out, before their winning streak, while the Giants were 75-70 before theirs.

Since there are two sides to the shut-out system, it's quite possible that backing the team handing out the shut-out went on to lose 6 games or more in 2011, hence the missing records. And what do we find - no less than four occasions when a team with a winning record inflicted a shut-out before going on a 6, 7 or 8 game losing run:

Baltimore Orioles 6-1, lost next 8
Detroit Tigers 12-10, lost next 7 
Oakland Athletics 22-20, lost next 6
Milwaukee Brewers 13-12, lost next 7

No wonder the records for 2011 are missing! As for the stakes you'd have been losing during those streaks, for the Baltimore Orioles they were 1.95, 1.95, 2.65, 2.7, 2.0, and 1.95 for their next six.

Suspecting now that the 2014 loss might not have been that early season sequence for the Brewers, I looked at the records for teams handing out a shut-out and then losing the next six or more. We can't get away from these Brewers because later in the season with a 52-26 record (seen that somewhere before as well) they lost 7 in a row. Since we can't trust the published records, I'll leave it there, but I did see the Braves also had a winning record that year, and has two losing runs of 6+ following a shut-out win. Hopefully readers get the point that while we may not know exactly what the criteria are, the method is flawed.    

Well that was fun. When I'm on my death bed, I have no doubt that I'll regret not spending more time on this.

Although these so-called systems are usually always scams, (a record of 113 – 0 is much sexier than one with 150 bets a year with an ROI of 2%) reading them can often trigger some ideas.

Why should a condition be the shut-out? It’s an easy number to see, but maybe a little more effort would be worthwhile. An unlikely scenario admittedly, but in theory a team could be shut-out after loading the bases in each of the nine (or more) innings.

Conversely, a team could avoid a shut-out by scoring one run off an error. Which team would you rather be backing in their next game?

Put in a little effort, i.e look at more than just runs scored and conceded, and you might just find an edge you can profit from.

And never buy from someone who doesn't know how to use apostrophe's:
His knee's what?  

No comments: