Across is the maximum I have seen from applying shoes. A very experienced stable once told me that 2 lengths is possible.

The figures across imply an improvement of 1.42 lengths.
13.54 - 11.96 =1.58
Minus Fitness (4%) = 0.16
1.58-0.16 = 1.42 Lengths

My advice is to expect on average between 0.3 and 1 lengths from applying shoes, don't go chasing 2 lengths. It'll very rarely happen, if Ever. Note: this effect may be positive or negative.

There have been many wild claims as to how much you can gain from the geld. Unless you have first hand evidence, don't believe it. In my time I have seen many, many gelds and across is first hand evidence of the biggest I have come across. A geld as big as this I believe is quite unusual.

Look at TT 3 and TT 4. The difference in lengths is 3.85 (7.21 - 3.36). Factoring in the fitness difference, 4%, which equates to 0.16 Lengths. This results in a gain of 3.69 Lengths.

My advice is to expect on average between 1 and 1.5 lengths from the geld, don't go chasing the 3 length geld, it really doesn't happen very often. The money spent going after unrealistic gelds will buy you many new horses.

NOTE: THIS DOES NOT FACTOR IN THE TRAINING RANDOM. IT COULD BE THAT TT NO.3 WAS AT THE UPPER END OF THE RANDOM RANGE AND TT NO. 4 WAS AT THE LOWER END. AFTER SOME RESEARCH INTO TT'S, I BELIEVE THE TRAINING RANDOM CAN BE UP 0.6 LENGTHS. THIS WOULD RESULT IN THE GELD ABOVE BEING A MINIMUM OF 3.1 LENGTHS AND A MAXIMUM OF 3.69

It is my personal opinion that blinkers range from minus 1.0 lengths to plus 1.0 lengths.

I have heard many claims of +/- 1.5 lengths and even a little more. The figures across may indicate that the blinkers added much more than 1.0 lengths to this particular horse.

Taking TT #140 and #141, the difference is an additional 1.57 lengths with the blinkers applied. This reduces to 1.45 when % fitness is factored in.

Taking TT #137 and #138, the difference is an additional 1.68 lengths with the blinkers applied. This reduces to 1.52 when % fitness is factored in.

Taking TT #138 and #139, removing the blinkers, didn't have the dramatic effect that the above two calculations would indicate.

Further testing confirmed that the effect of blinkers was more likely to be somewhere in the region of 0.8/0.9 lengths

This is a good example of variances and misconceptions due to the training random.

See across where further training indicates that the effects of the blinkers are indeed much less that first thought.