Analysis of betting every dog on the ML

I was asked to take a look at how betting on every game ML would turn out. So this is what I did:

I took every game since 91. I bet $100 on the dog ML. So if the bet lost, I lost $100. If it won, I won the ML payout. I used the following payouts for MLs (the spread is on the left, the ML on the right):

1 +107
1.5 +111
2 +115
2.5 +130
3 +145
3.5 +162.5
4 +180
4.5 +195
5 +210
5.5 +225
6 +240
6.5 +265
7 +290
7.5 +305
8 +320

and so forth.

So a $100 bet on a 5 point underdog would win you $210.

Here are the results:

You will see it's not really a good way to make money, as you would be putting up $25,600 to win about $585 on avg.

However, the interesting thing is to compare 96 and 95. In both seasons, 81 ML dogs won. However in 96 you would have lost $74 and in 95 you would have won $2,811.

The reason being in 96 the avg line that ML dogs covered was 4.2. Whereas in 95, the avg line that ML dogs covered was 5.1. So even that little change in the payout per bet amounts to a $2,900 swing...