My turn: MLB Overnight Betting

[dudeimstoked]

Sunday MLB
Minnesota +115
Seattle +133 (2)

Hopefully our luck will turn around.

[dudeimstoked]

I like the OVER 8.5 on Hou/ChC, but not at -125. If it drops to -115 or better, it is a play.

[New001]

[ QUOTE ]
I like the OVER 8.5 on Hou/ChC, but not at -125. If it drops to -115 or better, it is a play.

[/ QUOTE ]
Agree completely. I managed to get -110 this morning. [img]/images/graemlins/smile.gif[/img] Good luck.

Edit: And that's about as easy as these get. 6-5 after the second inning. [img]/images/graemlins/wink.gif[/img]

[dudeimstoked]

[ QUOTE ]
[ QUOTE ]
He sits day game after playing night game.

[/ QUOTE ]

Thanks for the tip. I'll incorporate this.

[/ QUOTE ]

Bonds played on Friday and we still lost :-(

[dudeimstoked]

Monday MLB
Florida Marlins -105
Washington Nationals +101
San Francisco Giants +126

My server crashed (some hardware problem) so I'll get that up and running sometime this week.

I am working on getting 1/2 Kelly bet sizing into my picks because right now it is not efficient as I am only estimating bets from the edge. Hopefully we'll have a sustained winning streak.

[dudeimstoked]

So I've been looking into incorporating kelly bets into my system, but the unit size will be thrown off like crazy.

For example, the Florida Marlins play at 1 unit under a 1/2 kelly bet would be a 4 unit play.

I need a better way to size bets, but this is probably too extreme.

[dudeimstoked]

Sunday Results
1-1-0 (-0.85 units)

Week Results
13-13-0 (+0.09 units; ROI: 0.33%)

YTD Results
20-20-0 (-0.03 units; ROI: -0.06%)

So it looks like all I'm good for is clearing bonuses...at least I'm not real negative i guess...

[dudeimstoked]

Added Monday Play
SF +126 (1 more)
This has been a 2 unit play all day, but I've been hesitant because Bonds has been sucking so hard. With a 0-20 Bonds, the line is not as favorable, but if he doesn't play, the line is actually still ok.

[Thremp]

dude,

Before you jump to Kelly you need a model that can determine your edge. Otherwise stick to flat betting. For example if you put together your model and it keeps giving out undervalued plays and you can beat the market, but your estimates are all over the board compared with actual movements, and you don't know how to refine it. Then you should probably just stick to flat betting. Once you can get it to accurately portray the odds of an event happening so you can determine your edge then you can bet Kelly safely.

[dudeimstoked]

My model actually does determine the edge given on any play. Right now I'm approximating bet sizes by only betting better than a 5% edge as 1 unit and 10% as 2, but I'm sure there's a better way because a 5% edge is better at +110 than +200.