Correctly Evaluating Results of a Simple System

[Cards37]

Hi all I posted in the today's tennis thread but I'm interested in whether I am tracking it accurately for evaluation so I thought I'd post here, hope this is the right place.

So I just wanted to start exploring the concept mentioned in one of Peter's videos. Keeping it simple I am laying favourites for $10 liability before the match starts, then looking at tennis trader for the projected price if the favourite serves first then goes down a break at 2-3 in the first set. I place an offsetting green up back bet set to keep in play at that price.

So far I have 34 results which I know is way to few for anything, but want I want opinions on is am I tracking results correctly so I can make a longer term evaluation then tweak as required based on good data, not being fooled by variance.

So obviously for starters I track the actual results - recording the lay amount, lay odds, lay liability, the exit back bet amount, back odds, whether the exit was matched, the match result, the profit/loss for the bet, and the commission paid.

From there I generate two data points - (a) the amount gained by "exit save" - the amount after commission of the successful back trade if the favourite goes on to win (calculated as trade profit after commission plus the returned lay liability) and (b) the "Missed profit if run" (lay amount less the traded out profit). If (a) is greater than (b) then I am assuming trading out at the nominated exit point has been an overall positive strategy.

Is it as simple as that? So far in an extremely small sample size the results are very unimpressive, hence why I want to test if I am tracking correctly to see if it's just variance. So far I calculate:

Matches: 34
Liability $303.05
Profit/Loss $-46.87

However, the "exit save" is $185.73, vs a "missed profit if run" of $63.08, which is $122.65 to the good. So am I just looking at variance here? Am I tracking things correctly, or have I made fundamental miscalculations? Grateful for any thoughts or advice.