Sunday 27 October 2024

The Buzz Around Killer Sports

Following on from the last post, we are indebted to Hkibuzz for providing more details about the upcoming changes at Killer Sports. He comments:

Hi Cassini. The free access to KS will be limited to 10 searches per sport per day. While it should theoretically be enough most of the time, it is still not enough on the days when you will want to do further research. I believe I will be a basic member, and hopefully it will mean the site will keep going for a long time. https://killersports.com/plans-premium
The linked page shows the following tiers and confirms what Hkibuzz stated:
As he says, the 10 searches per sport per day included in the Free tier should suffice for me on most days to identify what selections there might be for previously identified edges, but for researching new ideas the Basic tier might be required. Even though I am now retired, I think I can justify $19 a month if it becomes necessary.

One of the purposes of the manuscript is to trigger ideas for subscribers to carry out their own research, and it's gratifying to hear that at least some people are doing this. Hopefully the changes at Killer Sports won't prove to be show-stoppers for anyone. Tiers, not tears.

Thank you to Hkibuzz for your research and taking the time to comment. 

Killer Sports Pro Time

Weirimdi left a comment on the blog:

Hi there it is me again. What is your opinion on the new paid structure of Killer Sports? i don't want you to be positive or negative about it, but rather like to know if this will have an effect on the user structure and use cases of this great tool. I used the Db to experiment and to research. It was fun for me, but now i think a lot of silent researchers will leave the ship and this might lead to a decrease in researchers.
Killer Sports sent an email to registered users a few days ago, which included this paragraph:
Included in our tiered membership structure is a free access to the ability to run a limited number of queries / Simple Searches and some saved systems to satisfy our most casual users. The paid tiers expand from there to accommodate our avid users who want to frequently search the database and/or take advantage of our expanding number of valuable tools.
As I do not use the feature which allows you to 'save' systems, I'm hoping I will stay under the radar and be considered on of the 'casual users', but if the 'limited number of queries' is set too low, then I may be in the Pro tier, but at $39 a month, I would consider this to be excellent value. As my account is now showing Pro, I suspect I'm enrolled in the 30 day free trial.



As I wrote on the blog at the time, a little over three years ago that it looked like Killer Sports would be shutting down which would have been terrible, but fortunately the site was saved with more sports being added. It's not perfect as I have mentioned before. The data isn't 100% accurate but it's a wonderful resource nonetheless.

For the time being the site remains free to use but should that change I'll update you all here.

One subscriber asked me if a strategy of betting the Moneyline rather than Against the Spread (ATS) would be profitable. It certainly should be since the odds are aligned, but the beauty of ATS is that the ride is a lot smoother.

A few years ago I came up with a table showing the approximate 'true' Moneyline odds based on historical results from the spreads for American Football (F) and Basketball (B).

Like Killer Sports, it's not perfect, but I find it helpful and thought I'd share it here.

And no, that isn't a typo for favourites giving a point in Football.

Saturday 26 October 2024

Billions, Drama and Fading in the NHL

Tesla had a rather good finish to the week with the stock up 26% since Wednesday after their Q3 earnings report was released.  

Elon Musk made about $35billion on the day, so his net-worth spreadsheet updates last night would have been an enjoyable task. I assume I'm not the only one who tracks these things almost daily... 

My gains were a little less than those of Musk, and not enough to pull October into the green, but it at least made a dent in the red. 

The World Series got off to a dramatic start last night, with the game going to extra innings, the New York Yankees taking a one run lead in the 10th before the Los Angeles Dodgers hit a grand slam home run with two outs for a walk-off win.

It's unlikely that tonight's Game 2, or any game for that matter, will match that one for drama, but in sport you just never know.  

I mentioned the idea of fading popular teams recently, and Thomas suggested that in the NHL:
The popular team fading works the best for the Toronto Maple Leafs.

Overall, for the past ten seasons, backing the Leafs straight up or on the Puck Line in every game gives the third worst outcome of the 33 teams - Utah are new this season and have just six matches to their name so far. 

As Thomas added, when they are playing against a losing team they are top of the list with an official ROI of -12.7%

It's impossible to get a definitive answer on which team in the NHL is the most popular, in Canada but the most popular team there appears to be either the Leafs of the Montreal Canadiens who rank ninth worst in both scenarios. 

Monday 21 October 2024

Sports Equinox Starring New York

While the NBA season is about to begin, for the women the 2024 WNBA season is now over with New York Liberty dramatically beating Minnesota Lynx in overtime in Game Five of a best-of-five series for their first ever Championship. 


The WNBA systems finished the season with a 58.7% winning record ATS, a 17.8% ROI and a profit of 11.24 units from 63 selections. Since records began in 2011, the results are:
A similar set of results for the NBA season which starts tomorrow (Tuesday) would be most gratefully received. 

The World Series for baseball will be between the New York Yankees and Los Angeles Dodgers after the Dodgers clinched the National League title last night against the New York Mets. It'll be the 12th time the two franchises have met in the World Series although the first seven were when the Dodgers played in Brooklyn. The Dodgers have home advantage in the best-of-seven series with the first two games in Los Angeles and games 3,4 and 5 (if needed) in New York, before returning to Los Angeles for games 6 and 7 (again, if needed). 

October is always a busy month with all four major US sports active and there will be another sports equinox next Monday, 28th with an NFL game (Pittsburgh Steelers v New York Giants), Game 3 of the World Series in New York, 10 NBA games (including New York Knicks), and 5 NHL games (but sadly no New York team, so no single-city equinox to match Phoenix's achievement in 2001). New York City FC do have an MLS play-off game that day though, so not a bad effort by the Big Apple. 

After a slow start, the NHL systems are picking up with five wins from the last six selections:
As for American Football, it's not been the best of seasons so far. The less-followed CFL is slightly in the red with a record of 21-25-0, while the College Football systems are up a little with a 123-109-0 record. There are two NFL games tonight.

Friday 18 October 2024

Betting With the Heart

The NBA season is almost here, and a subscriber mentioned to me that:

Along the same lines I also avoid betting on the Lakers as this is probably the most hyped up team in the history of sport..
There's logic to this argument, and it's one I used in baseball one summer about 25 years ago when I was between contracts and would oppose the four most popular teams - Yankees, Red Sox, Dodgers and Cubs. 

The system didn't last long as another job came along, and it was with small stakes for entertainment rather than a serious effort to "bash the bookies", but my thinking was that casual bettors would back these teams and in theory there should be value on the other side. 

A few minutes after I saw the comment about the Lakers, ESPN published an article today on the topic of sports betting in New York  written by David Purdum.

The final paragraph reads:
"When the Yankees are going well, we'll most likely need their opponent every night," [Johnny] Avello, a New York native and longtime Las Vegas bookmaker, said. "They say, 'Don't bet with your heart,' but we all do."
Maybe not "all" of us, - at least not when it's sports betting (personally, politics is another matter) - but I took a look at the lines for the Yankees since the state of  New York legalised online gambling in January 2022, and backing them isn't as terrible an idea as I thought it might be. Of course it helps that they are a good team, winning 56% of their games, but the lines should reflect that over 500 games.

Blindly backing the Yankees would have lost you 3.1% of your bank, but when backed as a Road Underdog, you would have made 13%, the fifth highest return of five teams in double figures in this category. The markets tend towards a Home bias and a Favourite bias, which appears to hold true even if the away team is the Yankees.

Looking at the Lakers, and avoiding them does have merit, with a negative ROI of 11.2% since COVID, and a negative one as Road 'Dogs too in case you were curious.  In the playoffs it's a profitable strategy to back them, but as I seem to keep saying, the sample size is very small. 

A winning start with Thomas' two NHL selections both winning last night, but a losing start to the American Football with the Thursday Night game not following the script.

Thursday 17 October 2024

Inbox Nuggets for the NHL, NFL, NCAAF and CFL

Back in March, towards the end of the 2023-24 NHL season, I received an email from 'Anonymous fan' with an interesting idea for a system which had a promising track record supported by a solid hypothesis.

Hey Cassini,

Thank you for your work and your blog. I really appreciate everything you do. I have been messing around with killer sports since you introduced me to it and I have a lot of fun thinking about things that may matter for the sports I enjoy.

I think I may have found an intriguing one for NHL.

Since 2007- ROI +10.5% in 948 games

Since 2021- ROI +5.5% in 208 games

Since this season- ROI +12.9% in 51 games

It seems to be steadily profitable year after year, present a solid number of plays, and the games that meet the criteria are typically quite enjoyable to watch. I am not very good at math or statistics though so maybe it is too small a sample? Either way I was curious to get your thoughts.
Best,
Anonymous fan

I've said before that it's pitting your wits against others and the intellectual challenge of beating these markets that is in many ways more exciting than winning money and having fun with ideas like this is what it's all about. When we can make some of the ideas profitable, so much the better.

I did respond with my thoughts, but back in March I was about to head over to Arkansas for the Total Eclipse and with the NHL season wrapping up, I put this idea to one side and, if I'm being honest, forgot about it. Retirement was apparently a distraction, or I'm simply getting old. 

I was reminded of the system when Anonymous (I shall call him Thomas moving forward) wrote with an observation about an NFL system, and there in the e-mail thread was buried the NHL system.

With 17 full seasons of data dating back to 2006, (for some reason the data we need for 2017 is missing), the numbers are now:


My only concern is that the ROI is down for the latest five seasons, a trend to keep an eye on, but overall the results are good.

Included in my numbers are playoff games where the edge appears to continue.

The fledgling 2024 season has had just one selection to date, a loser, but with two qualifiers tonight it's another system I'll be adding to the Sacred Manuscript and sharing with subscribers along with Thomas' simple, but highly impactful, NFL observation.

After running the numbers, Thomas has essentially managed to obtain the same returns but from fewer than 56% of the selections. 

Side by side, here are the basic returns pre and post Thomas' modifier:

Excellent work, and two probable selections this weekend. It also appears to be effective in College Football with a similar improvement:

And in the CFL too, although with only a dozen games remaining it's a moot point for this season. Thomas deserves a medal.  

It's been an unusual NFL season so far, with just six games having a spread of more than a touchdown and extra-point, i.e. seven points.

At this stage of the season the average number of such games since 1989 is 18, and four of the six hotties this season have lost straight up, with only one covering the spread. 

Finally, and the topic of my previous post regarding shutouts in MLB playoff games is relevant again as we had another one last night in New York as the Los Angeles Dodgers beat the Mets 8-0. 

Wednesday 16 October 2024

MLB Playoff Shutouts

One of the recurring themes in the Sacred Manuscript is how the markets react to a team after they've been on the receiving end of what John Cleese might call a 'damn good thrashing.'


While the sample size in playoffs is always going to be small with fewer games to begin with and shutouts fairly rare, nevertheless when a team is shoutout (fails to score) and loses by five or more runs in the MLB, the market typically discounts them in the next game.

I may have mentioned this in 2014 when the World Series provided a good example. The San Francisco Giants beat the Kansas City Royals 5-0 in Game 5. The Royals got revenge two days winning 10-0 before the newly vanquished team (this time the Giants) bounced back to win Game 7 and the World Series. 

With such a small sample size, it's not surprising that the p-value isn't significant but with a good premise behind it, this is an idea that may be worth watching and of course, it paid off on Monday night with the New York Mets bouncing back from a 0-9 loss in Game 1 with a 6-1 win over the Los Angeles Dodgers in Game 2. 

It's encouraging that the Run Line returns support the theory, but it appears that the 'damn good thrashing' theory only applies when teams are shutout. 

Scoring just the one run in a big loss, and there is no carryover to the next game, nor is the theory validated in regular season games. 

Someone at Betfair actually settled the market after three innings as a Dodgers win at a time when they were trailing 0-6, but they corrected the error and brought the market back in-play after I pointed this out to them on Twitter. Mistakes happen, but the response was immediate which was as impressive as settling the market incorrectly in the first place was unimpressive. 
Game three of this serious is later tonight

Saturday 12 October 2024

Roller-Coaster, Tight Pants, and Bromley Car Parks

It's hard to believe that next month I will have owned Tesla ($TSLA) for seven years. I've written about this investment several times since 2018 and although it had a terrible finish to the week yesterday, it's still up some 940% overall for me, but down about 12% so far this year.  


I'm a little top-heavy on this stock as it currently represents more than 25% of my "play" account, but it's nothing like the position that Canadian Christopher DeVocht managed to achieve before things took a turn for the worse.
Christopher DeVocht, a carpenter by day, successfully traded Tesla stock and options during the COVID-19 pandemic, when a record amount of stimulus helped stocks soar and drove a retail-trading boom.

At the end of 2019, DeVocht had C$88,000, or about $65,000, with the Royal Bank of Canada's brokerage division. At its peak in November 2021, according to the lawsuit, DeVocht's account grew to C$415 million, or about $306 million.

But DeVocht and his professional advisors didn't cash out and instead lost everything as a brutal bear market took hold in 2022, leading to staggering losses for his heavily concentrated portfolio.
$65k to $306m to zero is quite an emotional roller-coaster. One commenter stated that:
Yeah I would never have made $300M. As soon as I hit $10M, I would have cashed out, set up an annuity or plopped it all into index funds and went fishing and mountain biking the rest of my life.

Easier said than done. While I'm a big fan of index funds and cycling, fishing - not so much, I suspect many people would struggle to call it quits on reaching $10M especially if that number is reached rapidly. It would be a nice problem to have, but very easy for that market driver of greed to take over.

With Kiké Hernández hitting a Home Run for the Los Angeles Dodgers as they eliminated the San Diego Padres last night, presumably while wearing his extra tight pants, and Elon Musk talking about car parks at the Tesla Robotaxi unveiling yesterday, I'd be missing a great opportunity if I didn't mention this post from a couple of years ago which addresses both car parks (Bromley Sainsbury's specifically) and baseball pant tightness. 

Monday 7 October 2024

MLB Totals, Playoffs, and Ice Cream

It seems a long time ago now, but the issue I mentioned in the last post regarding the line for the Green Bay Packers @ Tennessee Titans has now been corrected, and the winning bet will now show up in the 'official' results.

Peter confirmed my thoughts, and my betting, writing:

That Green Bay spread line from Killer Sports is wrong. I placed my bet at +3 only a few minutes before the start and was watching the lines across the board until the kick offs and it didn't move into favouritism. And covers.com reports on the result: "Green Bay covered the spread of +3."
Mistakes happen, but Killer Sports corrected this one fairly quickly, so no harm done.

The MLB regular season has ended and for the totals system which doesn't carryover into the playoffs, the all-time and 2024 season results are updated as below:


These calculations use the Pinnacle Dime line of -105 (1.952), but the system is historically profitable with the line as short as -123 (1.813):

Some of you may recall the UMPO system and while these low volume systems can be volatile in any one season and ROIs meaningless, historically the record is good, especially for qualifiers playing at Home and we have the first one for this season playing tonight. 
For some reason, Killer Sports no longer has the data for the first three seasons, but since 2007 the 'official' ROIs are 6.7% (all) and 14.2% (Home) - excluding the COVID season of 2020.

Due to the rule changes post-COVID, this system wasn't included in the Sacred Manuscript, but I feel confident now that it should be. 

On a personal note, our trip to the mountains was mostly very relaxing, but my wife was a little concerned when the campground was under an evacuation warning last Monday but personally, being surrounded on three sides by water (we were staying by a lake), I felt quite safe. In the end, no evacuation was necessary, and it was a very relaxing trip with no TV, and no computers (though I did have my phone with me). This meant plenty of time for reading and exercising, although the scales didn't have good news for me when I got back. The usual aphorism refers to doughnuts but adapted for me, it is true that 'you can't outrun an ice-cream'.

With now four years of personal health data, the clear trend is that I lose weight (hopefully fat) in the first quarter of a year (thanks to Dry January and New Year Resolutions) then gain weight every other month with the exception of October (thanks to Sober October, although there will be no fully sober October this year due to the trip extending beyond September by a few days).

It's often said that losing weight is 80% diet, 20% exercise but for me, calories ingested alone has a 91.1% correlation to weight loss, with exercise actually decreasing the correlation to 85.6%.

On a quarterly basis, if my average daily intake of calories is fewer than 2,230 I lose weight, as I do if I have fewer than 8 days a month with alcohol. It's that easy, but not very exciting.