_ 2023 November _

I don’t have the emails any more, but it looks like Keith was playing with the Reflex Unit

There is quite a bit of data in these files, definitely open them up`

OK, I will make a guess as to how the data is structured`

Here is a typical data set and it looks like he tried each of the experiments 10 times`

Here the test was to tighten up the reflex unit and see how much Extra Balls cost`

Test 1 = In Column B he started with 657 credits/coins on the machine`

Looks like the first Extra Ball cost him 1 coin`

The second Extra Ball cost him 16 coins`

The third Extra Ball cost him 20 coins’

_ etc _

With the Reflex tight and the worst conditions for features/awards`

Across the 10 experiments, it look like the cost of the first Extra Ball

1 , 5 , 5 , 4 , 19 , 4 , 41, 24 , 12 , 13

Wow`…if he diligently reset the Reflex each time and recorded these numbers accurately, that is quite a variance`

On top of that, the second Extra Ball cost an additional`

16 , 20 , 27 , 2 , 8 , 22 , 9 , 32 , 6 , 6

OK, equally bizarre, especially if this is the same machine in each test`

1-to-41 variance for the cost of the first Extra Ball`…Crazy

The average cost ~12-coins

If you don’t get it by the 6^th coin, maybe you quit`…I don’t know`

The average cost of the second Extra Ball is ~15-coins

With a 6-to-27 spread`…so obviously the Reflex doesn’t control this alone`

Looks like the Mixers are involved too`

As usual, these things are a Wonder

Lots to digest here`…Have fun`