← All tools

Autonomous Value

Whether the risky autonomous is actually worth more than the safe one, once its failure rate is counted. Uses the interval a small number of runs really supports, not the raw percentage.

Inputs

One row per routine you have actually run. Count every attempt, including the ones you would rather forget: dropping the bad runs is the fastest way to talk yourself into a routine that loses matches.

1
2
3

The bonus your game awards the alliance that scores more in autonomous. Take it from the manual. Leave at 0 if your game has no such bonus.

From scouting. Used only to work out how often you would take the bonus, so a rough number is fine as long as it is honest.

Results

Best expected value

Which routine

Its success rate

%

Which really means

%

95% confidence, Wilson interval.

Better than second by

pts

Per match, on average.

Break-even success rate

%

Where the top two routines are worth the same.

Every routine, ranked

Expected value is what the routine pays on average across many matches, not what it pays when it works.

RoutineWorkedRate95% intervalExpected ptsWorst case

How this is calculated

A routine that scores twenty points and works half the time is worth ten points a match. One that scores eight and works every time is worth eight. The gap is smaller than the headline numbers suggest, and it can reverse once the failure is worse than scoring nothing.

expected = rate × points when it works + (1 − rate) × points when it fails + chance of taking the bonus × bonus

Why the raw percentage is not the number

Seven successes out of nine is 78%, and it is tempting to plan around that. But nine runs is not much evidence. The Wilson interval, the same one the reliability tools here use, says the true rate is somewhere between roughly 45% and 94%. Both ends of that lead to different decisions, which is the honest answer: run it more before committing.

The worst case column takes the bottom of that interval. If a routine still beats the alternative there, the decision is safe. If it only wins at the top of its interval, you are betting on the good half of your own data.

Failure is not always zero

Fill in what actually happens on a bad run. A routine that ends up parked somewhere useful is a different proposition from one that ends up tangled with your partner, and a negative number is allowed if a failure costs you points or a penalty.

The bonus changes the shape

Where a game awards points for winning the autonomous period, a routine only has to beat the opponent, not maximise its own score. That can make a modest reliable routine worth more than an ambitious one, because it takes the bonus nearly every time.

Sources & assumptions

Autonomous lasts 15 seconds, which is published.

Point values, success counts and the opposing score are yours. The interval is a Wilson score interval at 95%, which behaves properly on the small samples a team actually has, unlike the textbook interval that gives nonsense near 0% and 100%.

Save this run, and compare

Keeps what is on screen so you can change something and see both sides of the change. Saved in this browser only, never uploaded.

Save this as evidence

Collects what you entered, what came out, how it was worked out, and anything the tool flagged, with a timestamp and a version so someone else can reproduce it.

This is evidence, not a notebook entry. It deliberately does not write your problem statement, your reasoning, or your conclusion, because under RECF rules an Engineering Notebook has to be the students' own work and no tool may generate or organise its content. Take the numbers, decide what matters, and write it yourself.