Sports handicapping advice and picks from a future professional sports handicapper.

Location: Lighthouse Point, Florida, United States

I am a corporate attorney with an MBA in finance. I've practiced since 1993. Currently I work for a high-tech company that focuses primarily on intellectual property development, technology licensing and venture capital transactions. In my spare time I am studying several so-called "advantage play" techniques as they relate to sports handicapping. I hope to someday pursue sports handicapping full time. Sportsbook managers, consider yourselves duly warned.

Sunday, July 24, 2005

Advice: Betting NFL "Teasers"

Let’s say that you are looking at betting the Bills -7 (-110) over the Dolphins. (Assume that you’ve evaluated the corresponding money line wager, but decided that money line isn’t offering any value and that it's better to play the point spread line.) You like the Bills in the game, and you think they’ll win, but you are afraid to lay the full seven points. First of all, you already know from reading past posts that seven is one of the dreaded “key numbers.” That fact alone makes it hard for you to lay -7 (-110). You’d much rather lay -6.5 (-110) if you can find it. What you’d REALLY like to do is bet the Bills -1, that would be great – and guess what - the sportsbook will let you do that, but there’s a catch. Let me explain.

The sportsbook will allow you to play something called a “teaser.” That is, the sportsbook will usually give you anywhere from 6 to 10 points to play with. You can use those points to add or subtract to any existing point spread. The catch is that in order to win a bet on a “teaser”, you have “tease” at least two games, and then you must get BOTH of them correct to win your wager. Using our example above, if you selected a 6-point teaser involving the Bills, you could play the Bills -1 … but you’d also need to select another game on that week’s schedule and also “tease” 6 points on that game as well. If the sportsbook was offering the Cowboys +2.5 over the Bears, you could then decide to use the second half of your 6 point teaser and select the Cowboys +8.5 over the Bears. Thus, your “2 team/6 point” teaser would look like this:

Bills -1 over Dolphins
Cowboys +8.5 over Bears

If you win one game and lose the other (or if you lose both games), you lose the entire teaser. If you win both games, your bet wins. Though payoffs on 2 team/6 point teasers vary somewhat, you should always try to find a book that pays off at even money.

As I noted, there are all kinds of these teasers available, involving all sorts of combinations of points and games. Luckily for you, you only need to remember one of them: the 2 team/6 point teaser. All of the other bets usually payoff at a rate that is very disadvantageous to the player (the vig is just too high), so you should avoid them. Leave the 10-point/4-team teaser to the knuckleheads who don’t read this blog; they are almost always horrible odds bets. The guys playing those odds should trade places with their wives at the slot machines.

Now that you know that 2 team/6 point teaser is the only kind of teaser you would consider playing; which of those 2 team/6 point teasers are the best? After all, it’s still very difficult to get both games right, even when you’re teasing 6 points on each game. There is a simple answer:

If you’re going to play 2 team/6 point teasers, find games that enable you to tease favorites DOWN through the key numbers (7,6,4,3) and underdogs UP through these key numbers. For example, Bills -7.5 vs. Dolphins and Cowboys +2.5 over Bears would be good teams to put in a teaser (if you like the Bills and Cowboys to begin with). They’re good because the 6 teaser points will tease the Bills -7.5 down through the key numbers (7,6,4,3), leaving you with Bills -1.5. Similarly, the 6 teaser points will tease the Cowboys +2.5 up through the key numbers of (3,4,6,7).

Historically, individual teams within 6-point teasers (not the entire teaser bet mind you, but each team within the teaser) have had the following results:

1. NFL home favorites of 7.5 to 8.5, when teased down, have covered 6 point teasers 79.8% of the time.
2. NFL home dogs +1.5 to +2.5, when teased up, have covered 6 point teasers 82.2% of the time.

Of course, historical stats don’t dictate future performance, but it can certainly serve as a good guideline.

So, the net/net of today’s lesson: If you’re going to play an NFL teaser, only consider playing 2 team/6 point teasers (preferably paying off at even money), and stick to teasing down home favorites of 7.5-8.5, or teasing up home dogs of +1.5 to +2.5

Good luck as always.

PS: There are also creatures out there called “reverse teasers” or “pleasers.” These work essentially the same as teasers, but in the opposite direction. That is, a pleaser is a wager that the favorite will cover more points that on the normal pointspread (or an underdog will require less points than it is getting on the normal pointspread) but I have not yet run the numbers on those, so I will save that discussion for a further post.

Thursday, July 21, 2005

Results: MLB 7-20-2005

Current YTD Record on Posted Selections:

MLB: 7 Wins, 6 Losses, 0 Pushes. Bankroll = +1.77 Units.

Wednesday, July 20, 2005

Picks: MLB 7-20-2005

A nerve-wrackingly big card for me tonight. Pretty soon I am going to start scaling back on MLB and buckling down on football. Here are tonight's selections:

Brewers (Sheets) vs. Cardinals (Mulder). Pick: Over 8 (+110)
Marlins (Moehler) vs. Diamondbacks (Webb). Pick: Under 9 (+110)
Royals (Lima) vs. Indians (Sabathia). Pick: Under 9 (+110)
Yankees (Small) vs. Rangers (Benoit). Pick: Under 13 (+105)
Athletics (Zito) vs. Angels (Byrd). Pick: Athletics (-105)
Dodgers (Lowe) vs. Phillies (Lidle). Pick: Dodgers (+145)

Current YTD Record on Posted Selections:

MLB: 3 Win, 4 Losses, 0 Pushes. Bankroll = -0.78 Units.

Advice: NFL "Key" Numbers

When betting NFL point spreads, you must be extremely mindful of what are called “key” numbers. That is to say, there are certain score differentials in NFL games that have a propensity to occur at statistically significant rates.

For example, from 2000-2004, games ended with a differential of exactly 3 points about 17% of the time and by exactly 7 points about 10% of the time. Other significant differentials were 10 (6%), 6 (5%), 14 (5%), and 4 (4%). Thus, there is a VERY good chance that the final score of your game will land right one of these key numbers … almost a 47% chance.

What does all of this mean to a bettor? In general, it means that there is a HUGE difference between laying, for instance, -3.5, -3.0, and -2.5 on a favorite. Similarly, there is a huge difference between taking +2.5, +3.0, and +3.5 on an underdog. SHOP, SHOP, AND SHOP SOME MORE TO AVOID BEING ON THE WRONG SIDE OF A KEY NUMBER. If you’re on the favorite, and you’re laying -3.5 (-110), you’re on the WRONG side of the 3. Similarly, if you’re on the underdog and you’re getting +2.5 (-110) you’re on the WRONG side of the 3 (from the other way.) That piddly extra half-point you gave on the favorite at -3.5, or that piddly extra point you didn’t get when you bet the underdog +2.5 proves EXTREMELY costly over the long run. It may not seem like a big deal at the exact time that you’re making the play (how important could a lousy half-point be?), but trust me, it is CRITICAL over the long run. Actually, don't trust me, run the numbers!

Again, if you like the favorite in a game, you should always try very hard to avoid laying -3.5 (-110): do your very, very best to get a -3.0 (-110) or, better yet, a -2.5 (-110). If you’re on the underdog, avoiding taking +2.5 (-110): do your very, very best to get a +3.0 (-110) or, better yet, a +3.5 (-110). The same advice applies for the half-points around the other key numbers: 7, 10, 6, and 4.

Memorize the key numbers 3, 7, 10, 6, and 4, and respect them at the betting window.

In some cases, and where it is permitted by the sports book, it can be mathematically wise to buy down off of a “3”; that is, instead of laying -3.0 (-110), you can pay extra vig and bet the same game -2.5 (-120). I'll discuss this tactic later on.

Results: MLB 7-19-2005

An even night on posted W/L last night, but a profitable night on units thanks to the Tigers pulling out a nice win. Look for more advice and possibly a few picks later today.

Current YTD Record on Posted Selections:

MLB: 3 Win, 4 Losses, 0 Pushes. Bankroll = -0.78 Units.

Tuesday, July 19, 2005

Picks: MLB 7-19-2005

Four selections in MLB for tonight's card, all from the computer model:

Marlins (Burnett) vs. Diamondbacks (Vazquez). Pick: Diamonbacks (-101), 1 Unit
Orioles (Cabrera) vs. Twins (Lohse). Pick: Twins (-103), 1 Unit
Devil Rays (Fossum) vs. Red Sox (Arroyo). Pick: Over 10 (-105)*, 1 Unit
Tigers (Bonderman) vs. White Sox (Contreras). Pick: Tigers (+125), 1 Unit

* Bad number; EVEN or better is now widely available.

Current YTD Record on Posted Selections:

MLB: 1 Win, 2 Losses, 0 Pushes. Bankroll = -1.00 Units.

Advice: "I'm pretty good, I usually get about 2/3 of my picks right ...."

Um, no ... you don't.

We've all heard handicappers selling picks on TV ("touts") claim that they're hitting 66% winners, 75% winners, etc. Our friends and neighbors often make similar claims. While hitting those kind of numbers is possible over a small sample, it is essentially mathematically impossible to do so over any significantly large sample size.

In football, picks against the spread usually pay off at -110; that is, lay $ 110 to win $ 210. In order to just break even at that rate, you need to pick winners at a rate of approximately 52.4%. If you can pick better than 52.4%** over the long run, and you avoid bad bets (like multiple game parlays and such), you can be a profitable handicapper. Most people cannot even pick 52.4% against the spread over any large number of games, let alone hit ridiculous numbers like 66% or 75%. Last year, the winner of the prestigious Las Vegas Hilton NFL SuperBook Handicapping Contest, a contest widely recognized as having the best handicappers in the world, hit winners at a rate of about 61% .... and that was only picking 5 games per week, FAR less than the full NFL schedule.

Let's assume that a 52.4% handicapper (better-than-average), made a pick in each game of the 256 game NFL regular season. How likely is it that this reasonably good handicapper could pick 2/3 of the games (about 171 games) correctly? Well, let's put it this way: binomial distributions tell us that there's literally about a 99.99989% chance that he can't. So, the next time your neighbor, your friend, or the guy on TV says that he's picking at 66% .... giggle to yourself. He might have done it over a weekend, but he could never do it over a season. If you want to amuse yourself, next time you are in Vegas walk into a sportsbook and ask if you can make a $ 10,000 bet and get 2:1 odds if you can hit 66% over the next NFL season and watch how fast they will agree to take your money. Larry Grossman, a popular Las Vegas radio personality, has a standing offer to anyone that he will pay 2:1 odds on any amount up to $1 million to anyone who can hit 70% winners against the spread over a football season. He has been on the radio for about 18 years; nobody has ever taken him up on it. "Sharp" handicappers know that anyone who could consistently pick at 60% or better would be Bill Gates rich ... they certainly would not need to be selling picks to you at $50 a pop.

So, do yourself a favor. First of all, don't ever believe anyone who tells you such nonsense. Secondly, set realistic targets for yourself: shoot for 55%, 56%, 57%, etc. Third, always make sure you're getting the best price for your bets because the better price you get, the easier it is to make money. Some sportsbooks offer "reduced vig" and let you bet against the spread at -105, or -107 ... if you can get those kind of wagers, you will need to hit LESS than 52.4% winners to break even. Also, some sportsbooks offer matching money bonuses, deposit bonuses, and other similar promotional offers. Those too help reduce the amount of winners you need to win money. Fourth, keep records. Be honest with yourself. Learn where you are making your mistakes and try to focus on them and correct them. Finally, remember that your biggest advantage against any sportsbook is that you can PASS on a game. Unlike the sportsbook, you don't have to have action on all 256 NFL games, you can just bet on the ones you feel you have an advantage on.

** The math is as follows:
1. Let x = loss rate.
2. Break even (0) = -110x + (1-x)(100)
3. x = .476 (approximately); 1-x = .524 (approximately)

Results: MLB 7-18-2005

A Marlins loss in extra innings turned a winning first day into a losing one, but bad beats are part of sports wagering. Advice and possibly a few MLB picks coming later today (7/19), after I run the model and check the lines.

Current YTD Record on Posted Selections:

MLB: 1 Win, 2 Losses, 0 Pushes. Bankroll = -1.00 Units.

Monday, July 18, 2005

Picks: MLB 7-18-2005

Here are three MLB plays for tonight (7/18) to get things started. My MLB plays come directly from computer software that I have developed; they will usually not be accompanied by the same level of analysis that will accompany a football selection.

Remember, always shop around to get the best price.

Tonight's plays:

The following plays all scored on my model as having a fair price of -110 or more. When I see "Even" and my model says "-110", that's a value play worth taking. Also, it's usually safe to assume that when it seems like the entire world is betting on one side of a particular play, I'll be on the other side of that play. I'm a big proponent of going against the "they're gonna kill 'em" mentality. That's what is happening so far on tonight's Yankees play ... everyone is on the OVER, which is just an added reason for me to put a unit on the UNDER. Good luck.

Marlins (Vargas) @ Diamondbacks (Gosling). Selection: Marlins (EVEN), 1 Unit.
Cubs (Williams) @ Reds (Claussen). Selection: Cubs (EVEN), 1 Unit.
Yankees (Brown) @ Rangers (Rodriguez). Section: Under 11.5 (EVEN), 1 Unit.

Current YTD Record on Posted Selections:

MLB: 0 Wins, 0 Losses, 0 Pushes. Bankroll (units) = $ EVEN.

Advice: Using Excel To Create an NFL "Season Wins" Tool

Several books are now posting "Season Win" totals lines on the upcoming NFL Season (e.g. Buffalo Bills: Over 8.5 Wins (+119), Under 8.5 Wins (-135)). For you novices, this means that if you bet the "Over", you will be wagering $ 100 to win $219. Conversely, if you bet the "Under", you will be wagering $ 135 to win $ 235.

Other than simply eyeballing these posted Season Win lines, how do you evaluate them for accuracy? Is 8.5 the right number to use? If 8.5 is correct, is -135 the "fair price" for an "under" bet? Finding value in lines is one of the keys to advantage handicapping.

There are several ways to construct an analysis tool for this. One that I like a lot was suggested by an acquaintance of mine over at Fezzik's Place (a link to Fezzik's Place is on my "Links" navbar.) Follow these instructions exactly in an Excel spreadsheet and you'll be able to build a very useful tool. If you follow it step-by-step, it's not as daunting as it may first appear. Before you use the sheet, you will need to estimate and assign a probability of the chance of winning each game of the 16 game season.**

1. In row 1 columns C through S contain the numbers 0-16 (so cell C1 = 0, D1 = 1, ... S1 = 16).
2. In column A rows 2 through 18 contain the numbers 0-16 (so cell A2 = 0, A3 = 1, ...A18 = 16).
3. In column B rows 3 through 18 contain the win probabilities for the 16 games (B3 = game 1 win prob, B4 = game 2 win prob, ...B18 = game 16 win prob).
4. Cell C2 = 1, cells D2 through S2 = 0 (as in, after 0 games there is a 100% prob of 0 wins and 0% prob of >= 1 wins).
5. Cell C3 = formula {+C2*(1-B3)} or the prob of 0 wins after 1 game = {prob of 0 wins after 0 games * prob of game 1 loss}.
6. Copy cell C3 formula to cells C4 through C18.
7. Cell D3 = formula {+D2*(1-$B3) + C2*$B3} or the prob of 1 win after 1 game = {prob of 0 wins after 0 games * prob of game 1 win + prob of 1 wins after 0 games * prob of game 1 loss}.
8. Copy cell D3 formula to cells D3 through S18.
9. The bottom row of the chart will then tell you the exact probability of acheiving the expected number of wins for each column's win total, based on your assumptions of the % chance of that team winning each individual game.

** (For now, you can just estimate these on the fly. In the future, I'll give you some ideas on how to arrive at a team's % chance of winning any particular game. After all, garbage-in still results in garbage-out. )

Once you have these probabilities from #9, you can construct moneylines, and then evaluate whether there is any "value" in any particular posted line. If you don't know how to do this final step (that is, convert %s into moneylines), I'll clarify it in a subsequent post. I'll also give you some ideas on how to arrive at a team's % chance of winning any particular game.


Welcome to Mike's Picks, a place with some good sports handicapping advice and (hopefully) some winning picks. Though most of my own handicapping time is spent on NCAA and NFL football, I do occassionally make selections on other sports - particularly on sports that are not widely bet and, consequently, may offer "inefficient" lines (WNBA, PGA Tour, etc.). Unfortunately I have no sponsors or affliations so you can be assured that neither my advice nor my selections will be influenced by any ulterior motive. Also, rest assured that I will have an "interest" in every selection that I post; I will not recommend a selection that I do not back myself.

When posting selections, I will always include my posted record to date, along with the number of units that I am up or (gasp) down.

Good luck. I hope you and your wallet enjoy my blog. The price is certainly right, though I hope you get more than you pay for.


Current YTD Record on Posted Selections:

0 Wins, 0 Losses, 0 Pushes. Bankroll (units) = $ EVEN.