[atakdog]

Dumb question, I suppose:

Why is playing the money line so unpopular (at least, I gather it is)? Who plays the money line, and under what sort of circumstances does it make sense?

I have a sense it may simply be tradition, in which case I would think there might be value there...

[ZeRo C@iDa*]

Dumb question, I suppose:

Why is playing the money line so unpopular (at least, I gather it is)? Who plays the money line, and under what sort of circumstances does it make sense?

I have a sense it may simply be tradition, in which case I would think there might be value there...

it was last Sundays ML on Giants that saved My @$s...

[SBR_John]

I love ML plays in the NFL and NBA. Not so much in NCAA. I think ML bets are smart bets in the NFL as long as your taking the dog.

[Scratch]

I think betting the faves on the ML is usually more risk than the reward is worth. However, I do remember reading in the paper about a wealthy businessman risking 800,000 to win 1 million on the 49ers ML over SD in the superbowl 94 season. We all know what an ass whippin that was.

[Scratch]

I should clarify he had to front the 800,000. He only profited 200,000 on his risk.

[Cyclone]

I have read that, instead of betting on an underdog of three points or less, you are probably better off betting on the moneyline. If the final score is really that close, your team has a decent chance of winning outright.

[rm18]

I should clarify he had to front the 800,000. He only profited 200,000 on his risk.

You would have had had to lay 15 dollars to win one on that game at least

[Ganchrow]

My answer to this question would be similar to the answer I gave to a related question on the comparison between baseball run and money lines.

I'd raise two issues:

1. Do we have reason to suspect that bookmakers systematically shade their lines in one direction or another (i.e., towards the fave or the dog)? To be perfectly honest, because I don't really bet football, and due to the scarcity of historical football money line data, I can't say I've looked too deeply into an answer to this question. Nevertheless, in Fixed Odds Sports Betting: Statistical Forecasting and Risk Management, Joseph Buchdahl presents a surfeit of compelling theoretical evidence (especially in the cases of rather lopsided markets) why bookmakers might choose to shade their money lines as to present better value for the favorite bettors.
   
   Furthermore, there's certainly evidence to suggest that many so-called "square" bookmakers have (at least in the past) shaded their spreads a not inconsequential amount so as to favor the bettor taking the underdog.
   
   Considering this all together, this tends to implies that there's probably no systematic bias which would work to favor money line dog relative to the dog spread on the same game.
   
2. To what extent and under what circumstances does betting market microstructure favor either spread or money line wagers? This is the issue I primarily addressed in link to the run line thread above. The conclusions I drew with respect to baseball wagers are valid in this case as well:
   
   
   As long as one have no specific reason to believe that there exists a pricing bias where the spread and money line are mispriced relative to one another then in general, for a given zero-vig edge, one should look to bet money line favorites and spread underdogs.
   As I wrote in the previously mentioned run line thread, "This ... will hold little weight for those that subscribe to the lottery mentality of sports betting, but for advantage bettors (whom we'd generally expect to avoid such ultimately detrimental staking strategies) it's clearly the superior option from the perspective of growth maximization and risk minimization."
   
   (I will note, however, that this will naturally be less valid right above money line break points. For example, if a money line market breaks from 20c to 30c at -190/+160, then markets at exactly that level or slightly will naturally be inferior to market slightly lower than level.)
   
   Let's look at tomorrow's (Sunday, Spetmebr 30, 2007) slate of NFL games at CRIS/Bookmaker. We'll compare the maximum possible expected growth (i.e., the E(G) associated with a full-Kelly bet) assuming a 7.5% edge at the no-vig (i.e., the "mid-market") price.
   

   ROT	DATE	TEAMS	SPREAD	ML	SPR WIN %	ML WIN %	SPR E(G)	ML E(G)	ML/SPR E(G)
   205	20070930	OAK	-110	+190	53.75%	35.58%	0.0376%	0.0266%	0.71
   206	20070930	MIA	-110	-230	53.75%	71.92%	0.0376%	0.1186%	3.15
   207	20070930	HOU	+100	-140	51.41%	60.42%	0.0399%	0.0900%	2.25
   208	20070930	ATL	-120	+120	56.09%	47.08%	0.0480%	0.0532%	1.11
   209	20070930	BAL	-110	-215	53.75%	71.00%	0.0376%	0.1771%	4.71
   210	20070930	CLE	-110	+185	53.75%	36.50%	0.0376%	0.0435%	1.16
   211	20070930	CHI	+105	-145	50.26%	61.39%	0.0438%	0.1019%	2.33
   212	20070930	DET	-125	+125	57.24%	46.11%	0.0576%	0.0557%	0.97
   213	20070930	GB	-110	-130	53.75%	58.34%	0.0376%	0.0679%	1.80
   214	20070930	MIN	-110	+110	53.75%	49.16%	0.0376%	0.0473%	1.26
   215	20070930	STL	-110	+650	53.75%	13.87%	0.0376%	0.0124%	0.33
   216	20070930	DAL	-110	-900	53.75%	93.63%	0.0376%	0.8324%	22.13
   217	20070930	NYJ	-125	-190	57.24%	67.74%	0.0576%	0.1102%	1.91
   218	20070930	BUF	+105	+160	50.26%	39.76%	0.0438%	0.0357%	0.82
   219	20070930	TBA	-125	+130	57.24%	45.17%	0.0576%	0.0580%	1.01
   220	20070930	CAR	+105	-150	50.26%	62.33%	0.0438%	0.1141%	2.61
   221	20070930	SEA	-110	-135	53.75%	59.40%	0.0376%	0.0787%	2.09
   222	20070930	SF	-110	+115	53.75%	48.10%	0.0376%	0.0504%	1.34
   223	20070930	PIT	-110	-270	53.75%	75.96%	0.0376%	0.2317%	6.16
   224	20070930	ARZ	-110	+230	53.75%	31.54%	0.0376%	0.0361%	0.96
   225	20070930	KC	-110	+525	53.75%	16.75%	0.0376%	0.0208%	0.55
   226	20070930	SD	-110	-650	53.75%	90.75%	0.0376%	0.7977%	21.21
   227	20070930	DEN	-110	+450	53.75%	19.01%	0.0376%	0.0230%	0.61
   228	20070930	IND	-110	-550	53.75%	88.49%	0.0376%	0.6231%	16.56
   229	20070930	PHI	+100	-150	51.41%	62.33%	0.0399%	0.1141%	2.86
   230	20070930	NYG	-120	+130	56.09%	45.17%	0.0480%	0.0580%	1.21
   231	20071001	NE	-110	-350	53.75%	80.85%	0.0376%	0.2826%	7.51
   232	20071001	CIN	-110	+290	53.75%	26.65%	0.0376%	0.0266%	0.71

   What we see without exception is that taking the money line favorite provides greater expected bankroll growth than the taking the corresponding spread. In the case of very large favorites, such as Dallas, San Diego, and Indianapolis, a full-Kelly money line bet (once again, assuming a 7.5% zero-vig edge) implies 22.13, 21.21, and 16.56 times as much expected bankroll growth as the corresponding bet on the spread.
   
   Similarly when betting on the money line underdog, in most (but not all) cases a bet on the spread provides higher expected growth than a bet on the corresponding money line. Where this not the case this is due to the significantly lower vig on the higher magnitude money line.

[atakdog]

My goodness; Thank you, Sir Ganch -- I have some digesting to do.

A related question, for you or anyone else: why would it be that often, when a book is taking unbalanced action in terms of number of bets (which data are easily available, in contrast to dollar amounts), the numbers of bets on the money line tends to be more balanced? This seems to be generally true (looking at the data available at SportsInsights), particularly for teams that are favored by between, say, 4 and 8 points (I haven't run a full analysis -- in fact, I'm trying to decide whether it would be worth the effort -- but that's my first impression).