|
| |
We don't believe there is a discernible difference in the relative chances of
predictability between our 3rd best game to our 4th best game. We do believe
that certain groups of games will have a relatively higher chance of being
successful versus others; however, we see that line as a very slim line, one made up
of fractions of a whole. We also don't see that variance in predictability deterring
us from feeling as confident about our 7th or 10th game as we do our #1 or #2
(to the degree that we would advocate not playing it at the expense of a higher
play on a higher pick).
Knowing that it takes 52% winners to break even (just trust us on this one), that
the typical person spreads between 3-6 games a week, and assuming the average
better wins at rate nominally accepted to be 50%-49%. So in a 3 game week,
anything less than 1 loss constitutes a losing week, and given a 5 game week, any
more than 2 losses constitute a losing week. Seems like a tall order, huh?
|
|
| |
|
|
|
 |
|
| |
Going back to what we stated above, even if we were consistently returning 60%
winners, in a 3 game week (assuming all 3 games carried the same relative value and their was a penalty of 10% for a loss), my 60% would land me somewhere between 1 and 2 losses a week, and in the 5 game example, I'd consistently be at 3-2, for a return of less than 1 unit (assuming that was a standard play). So even if I am an exceptional handicapper (and by nearly all accepted standards, 60% qualifies as exceptional), I still am scratching and clawing each week to return relatively low absolute returns based on my investment. Consider the 5 game scenarios: Obviously a 5-0 week is great, a 4-1 week would return +2.9 units, not bad, a 3-2 week (60%) returns less than a unit, while a 2-3, 1-4, and the dreaded 0- 5 are all losers. Throw out the 5-0 and 0-5 as the exception rather than the rule, and out of the 4 remaining outcomes, only 1 of them returns anything of substance, this all from a handicapper at the top of his game, and industry. |
|
| |
We prefer to employ a slightly different strategy than most. I'm not going to come out here and tell you I've got only 4 plays, and they are GOYs, GOMs, Dog of year, whatever. I will consistently evaluate the week based on the same criterion set, and determine how many games are strong enough to garner a play. From that we will group them relative to their predictability. We won't have flashy stars or"of Year" tags. We will simply cluster them by relative strength. We employ a very straight forward approach, a #1 play is the strongest play we have (but we typically have 3-6 of these a month), followed by a #1/2, #2, #2/3, #3, #3/4, #4 and"others". This results in anywhere from 6-14 games spread a week. However, the relative strength of the #1 play to the #4 play is a little over a unit. We employ the following strategy: |
|
| |
#1 – 1.8 units
#1/2 – 1.6 units
#2 – 1.4 units
#2/3 1.2 units
#3 – 1 units
#3/4 - 0.80 units
#4 - 0.60 units
Other – typically not advocated as a play, but if so no more than a 0.50 unit play. |
|
| |
This method also requires a re-evaluation of what a comfortable bankroll is, and
based on the higher quantity of games spread, what a re calibrated unit is. I
won't get into that here, but it is a pretty simple calculation, based on what you
can comfortably afford to lose in a season divide by the weeks in the season you
would spread it over, divided by the above average units of games per week, than adjusted for some percentage based on a worse case win scenario.
The reason behind the higher number of games and lower range b/w levels is
what we alluded to earlier. Based on our system, we may feel that a number one
qualifies in a couple scenarios that have consistently proven to return and 75%,
73%, 68%, and 64%. |
|
| |
However, our #4 pick may qualify in some (if not all) of those same scenarios, but may have a slightly lower coverage percentage, or slightly higher variance in model alignment. Or we may feel the #1 play, based upon further analysis, has some stronger situational strength. At the end of the day we feel they are both solid plays, and the relative increase in predictability in the #1 play versus the #4 play is very minor, perhaps say 2-4%. We'd prefer to not dilute a sustained winning percentage by high losses when that 2-4% was inaccurate. We prefer to let a consistent winning percentage over a steady stream of games produce higher weekly unit gains, and with relatively less stress. Who wants to go 7-0 early in the day only to sweat out all you have made that week on the last game which some one told you is a "cod lock"?
Remember, THIS IS SUPPOSED TO BE FUN!!!!! Set a reasonable bank roll, if you
can't afford to lose the money, it will never be fun no matter what you win,
establish your own predetermined criteria based on your bank roll and relative risk tolerance, and enjoy college football, it is the best game there is. |
|
|
|
|
|
|
