With the 2019-20 Bundesliga and Bundesliga.2 regular seasons finally complete, it's time to update the Booming Bundeslayga post from earlier this month.
Tuesday, 30 June 2020
Bundesliga 2019-20 Notes
Sunday, 28 June 2020
Leaning Green
Back to politics for this post, and some interesting comments from Paul Motty, aka Political Gambler regarding the 7% implied probability that the next president will be neither Biden nor Trump.
Both candidates are up there in years, and both have - not surprisingly - shown some decline from their prime years, and while money coming in for Pence is not a surprise, the same for Clinton is a little unexpected.
As mentioned previously, the five states to keep a close eye on, and in order of importance, are Florida (29 votes), Pennsylvania (20), North Carolina (15), Arizona (11) and Wisconsin (10).
Michigan (16) was a "toss-up" state a few weeks ago, but as I wrote here, has now moved to the "Leans Democratic" category. A similar move was made this week by the Centre For Politics web site for Pennsylvania.
Texas may yet move to "toss-up" after polls in that state show the race there to be neck-and-neck.
Florida has been hit hard in the last few days by a huge increase in new cases of COVID-19 and the Republican governor there hasn't managed the crisis well at all which may hurt them in November.
There is still a little over four months to go until the election, and a lot can yet happen, but for now, anyone following along is in good shape. Leaning Green you might say.
Latest odds on Betfair are Next President Biden at 1.68, Trump at 2.92; Popular Vote Winner 1.19 / 8.2 respectively; Winning Party Democrats 1.56, Republicans 1.76.
Friday, 26 June 2020
Long Division
How the 'empty stadium' factor will influence the 'home advantage' in baseball this summer remains to be seen, but the biggest advantage (batting last) of playing at home will not be changed.
Aficionados of the game might be interested to learn that in the early days of baseball:
the home team could chose to bat first, and would sometimes do so to gain a tactical advantage (for example, by trying to rattle an opposing pitcher by forcing him to pitch in front of a hostile crowd before he had had a chance to adapt). Most managers realized however that the tactical edge of batting last in the final inning was much more precious than any short-term gain from batting first.Historically the Home side wins about 55% of games in baseball.
While the 60 game schedule is not yet finalised or released, it's almost certain that more games than usual (at least as a percentage) will be Divisional games, so in preparation I took a look at the last few seasons to see how the market behaves for matches in this category.
Readers of this blog will know that in most of the US sports leagues, Divisional games have their own characteristics with the teams playing such opponents several times a year in the case of baseball and ice hockey and home and away every year in the NFL where 50% of regular season games are Divisional and the markets historically under value the road team.
But back to baseball, and here are some statistics going back to the 2012 season for 'hot' favourites, which for the purposes of this discussion are teams priced at 1.5 or shorter, which is -200 in American odds.
As readers will know, hot favourites have been unstoppable since 2012 with an ROIs of 6.3% and 5.7% on the Money Line and Run Line respectively. The average line on these selections has shortened from -224.8 (1.45) in 2012 to last season's -256.1 (1.39) while the number of selections in the same period has jumped from 202 to 474. That should equate to at least 175 selections in the shortened season ahead.
In Divisional games, the ROIs are 6.3% and 7.8% overall, but when the Away team is a hot favourite, the numbers jump to 15.8% and 16.2%, and in the American League to 17.9% and 18.4%.
The findings for the T-Bone System are less exciting with the ROIs since 2012 dropping in Divisional games from 7.1% and 7.2% to 3.9% and 0.9%.
Thursday, 25 June 2020
MLB Sprint
Some good news for the Summer with baseball looking like it will be back, albeit for just a 60 game schedule and with a couple of significant rule changes.
The Designated Hitter rule will be universal and in extra innings, each side will start with a runner on second base.
Baseball has experimented with this in the minor leagues, and according to @JJCoop36:
The regular season becomes a sprint rather than the usual 162 game marathon, so there will be differences in how managers approach these games. Loss aversion is likely to be higher, as is the probability that many teams will stay in contention for the play-offs much later in the season than usual, simply because there is less time for them to drop out of contention.
The planned schedule will see the season open around July 24th and end on September 27th, with the playoffs and World Series immediately thereafter.
It also looks likely that within each region, teams will play the majority of games against others in their own division instead of adding more inter-league games. One report suggests that each team:
"will play 10 games apiece against its division-mates for a total of 40 games. The remaining 20 games will come against the other division, though it's unclear if it will be an even breakdown of four games against each inter-league opponent."
Readers will know that backing hot favourites later in the season is a successful strategy with an overall ROI of 6.9% on the Money Line and 7.2% on the Run Line over the past eight seasons and close to 1,250 matches. For road teams, the ROIs are in double figures over this period, but with empty stadiums, perhaps the market will be less inclined to over-bet the Home teams? I have a feeling that more competitive games as the season winds down may reduce the profitability but something to watch.
With the increase in Divisional games, systems that take this parameter into account should have relatively more selections.
Also worth considering is that implementing the Designated Hitter rule in every game should see more runs being scored, as should the "runner on second" rule.
If the markets don't adjust, or if they over-adjust, there will be value to be found in this area, but readers of this blog will know where value in the Totals markets has traditionally been found. No one has ever seen the DH rule applied in a NL Division game before, so this is new territory. With change comes opportunity.
Tuesday, 23 June 2020
Look Away For Favourites
With another "Local Derby" draw this weekend on Merseyside, I thought I might share another observation, although I may live to regret this.
The numbers below are from 2000 to the suspension of play this season, as I haven't updated my spreadsheets yet with this week's results, but the trend is clear.
Basically in close matches, i.e those where the teams' win probabilities are within 25% of each other, the value on the Draw is much higher when it is the Away side who are favourites, or when the two teams are priced exactly the same.
The Draw on Pinnacle is currently 3.54 which is great value in a "toss up" matches with only 23 previously this high in almost 20 seasons with an ROI of 82%.
Sunday, 21 June 2020
Local Derby Factor
It's the Merseyside Derby today and Derby games traditionally are good value for Draw bettors.
Since 2000 in the EPL, the return from blindly backing the Draw in Derby using prices adjusted to an over-round of 103.7% is 1.1% which may not sound like much, but it compares well with non-Derby matches which would have lost you 4.47%.
For the record, while games such as Liverpool v Manchester United and Crystal Palace v Brighton and Hove Albion are often described as 'Derby' matches, the term is used as in 'local Derby'. Big rivalry games is another category.
The "Close" matches means matches where the teams are within 25% of each other in Win probability, and "Toss Up" is where teams are within 10% of each other.
Little evidence yet that Home Advantage has vanished in the EPL with the three teams winning Away so far all playing at clubs in the bottom four of the table.
Meanwhile things aren't looking great for baseball this summer. While the league and players argue over how many games to play and how much they should be paid, the two states where "Spring Training" takes place (Arizona and Florida) are both seeing surges in the number of new cases, the result of poor leadership at the state level with Republican Governors opening up the state prematurely, going along with the argument that the "cure can’t be worse than the disease”. As the disease is killing people, it's hard to understand how the cure could be any worse.
As a result, teams have announced they will train in their own stadiums, potentially costing Arizona and Florida a ton of money and hopefully Republicans a ton of votes. With cases on the increase, MLB should probably just call it a day. Assuming, and it's a big assumption, that the US get the first wave under control, the second wave is likely to arrive around World Series time, hence the reluctance of the players to extend the season. Spring has gone now, so the traditional pre-season warm ups would be "Summer Training" but to me, it looks like they are running out of time to strike an agreement, and the conditions don't support a return any time soon anyway.
With conditions worsening in Florida, the NBA's plan to resume play at the Disney World Resort in Orlando might be re-considered, and the NFL training camps are due to start next month. It's hard to see how that season will be anything like normal.
I've not checked the prices today, but the Biden and Democrats to be the winning party bets can't have been hurt by last night's rally in Tulsa, where Trump kicked off his campaign. With ticket demands over one million announced several times on TV last week, Trump was no doubt very excited and scheduled to speak to 19k inside and another 40k outside but it seems he was punked by teenage Tik-Tok users ordering tickets they never intended to use, and the actual attendance inside the arena was 6,611. The outside crowd was zero, and that speech was cancelled. He's finished, and he surely knows it.
Saturday, 20 June 2020
Michigan Leans Blue
The US Presidential Election is three weeks closer (now 136 days away) than it was when I wrote this last month about the so-called "toss-up" states:
In Florida, polls as of yesterday have Biden leading 49% - 46%. Clearly with 29 votes, this is a must win state for the Republicans, as if they lose here, the Democrats would need just one of the other swing states to push them over the 269 votes needed.This means that with 248 electoral votes in the Solid, Likely and Lean categories, the Democrats would need only 23 of the available 86 (27%) electoral votes from the Toss Up column.
Pennsylvania has Biden well ahead by 49% - 41%, although the polling data here is eight days old, while Michigan and North Carolina have Biden ahead 49% - 43% and 46% - 44% respectively. Biden also leads in Arizona (47% - 43%) and Wisconsin (47% - 41%) which all means that in a fair election, anything close to evens for the winning party to be Democratic seems great value.As of yesterday, the Cook Political Report (an independent, non-partisan newsletter that analyzes US elections) updated its web site to reflect that Michigan, with its 16 Electoral College votes) is no longer a toss-up state, but has moved into the "leans Democratic".
A win in Florida and it should all be over, and the latest polls there have Biden ahead by 50% - 43%, 51% - 40% and 50% - 40%.
Hardly a surprise then that the odds on Joe Biden have dropped sharply in three weeks from 2.17 to 1.82.
For the Democrats to be the winning party they have dropped from from 1.96 to 1.7 and for the Popular Vote winner to be Biden from 1.34 to 1.2.
Last Wednesday, @BradAllenNFL tweeted:
I'm starting to think Biden at 1.88 might be one of those Mayweather v McGregor once-in-a-lifetime type betsHe may not be 1.88 any more, but at 1.82 is still value in my opinion, and one reason why is summed up by the Ladbrokes Politics Twitter account.
Friday, 19 June 2020
Sound of Silence
There are a number of people jumping on the "no crowd / no home advantage" bandwagon, although it is very early days in the experiment and the number of trials is still very low.
He suggests that five teams are worthy of "consideration" whatever that means adding that "all five are realistic shots. But what’s remarkable is every single one of them is still available at an odds-against price. The prices are assuming home advantage still exists."
The noise of the crowd influenced referees' decisions to favour the home team. It is suggested that referees' decisions are influenced by the salient nature of crowd noise, the potential use of heuristic strategies, and the need to avoid potential crowd displeasure by making a decision in favour of the home team.The results found that:
The presence of crowd noise had a dramatic effect on the decisions made by referees. Those viewing the challenges with background crowd noise were more uncertain in their decision making and awarded significantly fewer fouls (15.5%) against the home team, compared with those watching in silence.But this is nothing new. What is new is the possibility to more accurately evaluate this advantage, although I suspect it has eroded somewhat since 2002 by the implementation of goal-line technology, VAR and modern schedules as I have written about before. I suspect this will be a returning topic over the next few weeks.
Wednesday, 17 June 2020
Over-round 2019-20
For comparison, it was 102.8% for August to December, and last season was 102.4% following seasons of 102.1% and 102.0%.
A concerning trend for those who take their sports investing seriously, although still a huge improvement from the 2000-01 days of 112.5%! Did anyone other than the rare and lucky accumulator winner ever make a steady profit against this margin?
Such variations over the years make direct comparisons from one season to another irrelevant, but we can standardize the over-round and work with adjusted prices. It's not perfect but it gives a more accurate comparison than using the raw numbers.
For example looking at the returns from backing the Draw in the EPL since 2000-01, using the raw numbers (and prior to Pinnacle's Closing Prices from 2012 I used 2000-02, William Hill from 2000-02, and Bet 365 from 2002-12), backing the Draw in every match would have cost you 555.77 units.
Using the imperfect over-round adjusted to the current 103.7%, the loss from those 7508 matches is 290.12 units, which readily illustrates the impact the over-round has on returns.
The table below shows the results for all blindly Backing outcomes:
Fortunately it is blindingly obvious that some bets are always just terrible value, and if you stay away from these, you are at least in with a chance of being profitable.
Sunday, 14 June 2020
Booming Bundeslayga
With the German leagues back in action for a while now, and prompted by an inquiry from Ioannis, I thought I'd update my Bundeslayga System results.
In Bundesliga.2 the numbers are 12.05 points (8.3%) and 4.63 points (21%) respectively.
Friday, 12 June 2020
Drawing To A Close
Church of Betting is one of the better betting related blogs out there, and the latest post is on a topic close to my heart - Draws.
Nenko looks at how the strategy of backing the Draw in 'close' games has played out since the basic idea received attention with the publication of David Sumpter's book "Soccermatics" in 2016.
Sumpter's idea was to back the Draw in matches between the Big 6 clubs of the English Premier League, but to my mind this strategy was flawed because in any given season, one or more of those Big 6 may not be that big at all, and it seemed to me that a better strategy might be to look at matches between relatively evenly matched teams, regardless of status, where the implied win probabilities were close.
For example, in 2018-19 Manchester City were priced at 1.27 to beat Tottenham Hotspur in a Big 6 match and backing the draw in this match made little sense while ignoring matches between non-Big 6 clubs where the two win prices were perhaps identical.
'Close' is a relative term of course, but Nenko looked at matches where the difference was less than 10%.
To remind you, the original idea recommends backing the draw in the Premier League in games where the difference of the winning probabilities of the two teams as implied by the odds (after adjusting for overround) is less than 10%. In other words, the market implies the teams are somewhat evenly matched in strength.One flaw in Nenko's analysis is that he uses Pinnacle's Opening prices to determine qualifiers.
A quick note: there is a small difference to my initial article since I used maximum odds among all bookmakers back then. Now I am only using Pinnacle’s opening price.The strategy I recommend is to use the prices as close to kick-off as possible for maximum accuracy in the win probabilities of the two teams and so the Closing prices actually measure the effectiveness of the strategy as markets are not as efficient when they open (possibly weeks in advance) as they are close to kick-off.
If we use Pinnacle's raw Closing prices as our filter, over the last four seasons, including this incomplete one, there have been 108 qualifiers, with a profit of 9.42 points, and ROI of 8.7%.
When I say 'raw' I mean prices that haven't been adjusted to account for the over-round. While it should be preferable to calculate the actual implied win probabilities and use these to determine qualifiers, the extra work seems to add little value as using this method there have been 105 selections, with a profit of 8.8 points, and an ROI of 8.4%.
So in the EPL, results for this strategy when correctly applied appear to be holding up, even though the genie is out of the bottle. Fortunately punters generally have short attention spans, and after a short-term dry spell will forget about the long-term profitability of a method and move on.
Looking at the 2019-20 results so far, I imagine several followers gave up at the end of January after one winner in 12 matches, but these losing runs have to be expected when it comes to backing the Draw.
Note that matches in bold are those where the difference was less than 5%.
Similarly in La Liga, which also came under Nenko's scrutiny, the results over the past (almost) four seasons are also profitable, with an ROI of 6.7% from 96 selections.
One noticeable change in this league is that prior to 2016-17, the Draw in 'close' matches was never shorter than 3.1.
However, in the (almost) four seasons since, the Draw has been priced shorter than 3.0 on 14 occasions, and in no fewer than nine just this season. So far!
Backing the Draw at sub 3.0 is rarely value, and you can improve your ROI by excluding these qualifiers.
Thanks to Nenko for giving me something to write about as the pandemic plays out.
Sunshine Follows >5% Declines - Usually
Who needs baseball, I rhetorically asked at the end of my last post. After today's beating in the markets, it seems that I do. Some slow and steady T-Bone profits are far easier on the psyche than the losses incurred by a 5.9% one day loss in the markets.
This was only the 28th time since 1952, when the S&P 500 converted to a five-day trading schedule, that the index has tumbled by at least 5% in a day with five of those declines have been in the past three months alone.
The good news though is that Market Watch reports that "declines of this magnitude have historically been followed by sizable rebounds in the days, weeks and months to follow".
Investing is a game of endurance, with the biggest gains accruing to those that can most withstand the pain.
Tuesday, 9 June 2020
Boeing Meets Tesla
Boeing are now +57.56% after just one week!
I don't think I'll be selling anytime soon either, with this Barron's headline:
Boeing Stock Is Surging Again. It’s Looking More Like Tesla These Days.
Tesla is the +202.37% investment seen on the left, so there seems to be plenty of runway left for Boeing.
Currently at 230.5, Boeing was at 347.45 on February 12th, so it really does have some upside with Seaport Global Securities today recommending a Buy and looking at a $277 price target.
Normally I'd be quite pleased with the 3.11% on Pfeizer, another oversold stock purchased at the same time as Boeing last Monday, but performance is all relative, and relatively it has been disappointing.
Both are components of 30 stock Dow Jones Industrial Average which was up another 1.7% yesterday.
Another major US index is of course the NASDAQ, which is heavily weighted towards tech and healthcare companies, both of which have performed well during the COVID-19 pandemic.
As a result, this index set a new record high yesterday, and the 76-day return to a record high is the second fastest rebound on record (a new high in a new bull market), after the 65 days ending on April 9, 2009.
Who needs baseball!
Saturday, 6 June 2020
Splendid Hibernation
As the years have gone by, the impact to my bottom line from trading and sports investing has diminished in percentage terms, and lately in actual terms also.
Other than some unsettled trading on Politics mentioned in my last post, May saw no trading or betting activity at all, with all my sports and markets continuing their enforced hibernations. If you were brave enough to follow me on the US Presidential Election bets, you're in the money right now with Biden currently the odds-on favourite, and the Democrats around 1.85 to be the winning party. Keep an eye on those swing state polls, but I'm not seeing any reason to lock in profits at this time.
...where 22 of the 30 NBA teams are expected to participate in Orlando. That will include the 16 current playoff teams as well as the New Orleans Pelicans, Portland Trail Blazers, Sacramento Kings, San Antonio Spurs, Phoenix Suns and Washington Wizards. Teams will play an abbreviated version of the regular season that will consist of eight games, as well as a play-in tournament for the eighth seed in both conferences. The tournament will only happen if the No. 9 seed finishes within four games of the No. 8 seed, in which case the No. 9 seed will have to beat the No. 8 seed twice to earn the final playoff spot in their conference. The playoffs will be a best-of-seven series in each round, with the NBA Finals concluding no later than Oct. 12.