David: As briefly as possible:

We offered to answer every one of your questions as long as you answered ours. You said no. You said you were afraid your wife would take your baby away from you. Now you are complaining that we don't answer your questions. One can't make this up, and one can't imagine the effrontery required to ask MORE questions, questions based on false and hilarious premises, to boot.

You have changed your percentages, which I called "idiotic." But your new percentages are only slightly less idiotic. If Ragozin's variants are steady and JB's yo-yo up and down "ON ANY GIVEN DAY" (as you wrote) and do all the other crazy things you've said they do, the numbers will not be in sync 70-85 percent of the time. Now you're backing off of "any given day," and you're even changing the meaning of "yo-yo," for God's sake.

You ignored my first paragraph, which means you believe Ragozin's numbers are invalid.

You have changed positions constantly, backed off numerous times, and are now left with this:

"JB has a horse at CT running straight 5s. On Ragozin, she runs 10s-12s. As long as that horse stays at CT, she will continue, I think, to run tight numbers [on TG]. Only when horses bounce around to different circuits with different scales is there a risk that their TG sheet will collapse."

Completely wrong. What matters is the accuracy of the numbers. Let's say there is NO circuit changing. If you start out with artificially tight cycles (inaccurate numbers, in other words), horses to whom you gave 5s will frequently lose to horses to whom you gave 7s. When this happens, your cycles will automatically get looser.

I will make it even simpler for you. 5F race. Only sprint of the day. One length equals about one point at this distance. Five-horse field. Here are their figures going into the race:

Horse 1: 5555555555.
Horse 2: 6666666666.
Horse 3: 7777777777.
Horse 4: 8888888888.
Horse 5: 9999999999.

Cycles cannot get any tighter than these. Then they run. Ground loss and weight are even. Horse 1 wins by one length over horse 2, who finishes one length ahead of horse 3, who finishes one length ahead of horse 4, who finishes one length ahead of horse 5.

What should the variant be for this race? Easy, right? A number that gives horse 1 another 5, horse 2 another 6, etc.

Now let's say the numbers we had given these horses GOING IN TO today's race had been inaccurate. Let's say horse 5 had really been running 5s, horse 4 had been running 6s, and so on. The order of finish will be the reverse of the way I showed it above. What's your variant now? What happens to the tightness of the cycles?

Now try something less far-fetched: Pretend that all the figures were accurate except the most recent number for horse 5. Say we had it one point slower than it should have been. Say he pairs it and DHs for fourth. What happens now? You throw him out of your projection (giving four other horses the numbers they usually run), and his cycle gets slightly looser.

Inaccurate figures = "implausible" results = a loosening of the cycles in order to incorporate the "implausible" results. That's another way of saying that fast horses tend to beat slow horses and if your numbers don't show which horses are faster and by how much, your cycles will get progressively looser.

Patent: "We've gone round and round on empirics vs. belief and I don't expect you'll ever come around on that one."

I think you mean "empiricism," David. In truth, TG and Rags numbers are based on a combination of experience/observation and theory. (By the way, who makes Ragozin's Maryland numbers? Who makes his California numbers? Who makes his Illinois numbers? What are the names of these trackmen who are counting the raindrops and estimating evaporation? In your system, they become extremely important.)

Patent: "I would wager that [25-35 percent of the numbers have to be inaccurate] before the sheet just didn't make any sense."

You are changing the terms of the debate, which is not a bad idea when your previous position is indefensible, as yours was. The question is whether tight cycles are sustainable if the numbers are inaccurate.

"What percentage of the horse's numbers would have to be wrong for the tight ranges to collapse?"

See above. One bad number can loosen the cycle, and there is a cumulative effect. Bad numbers do damage in future races, producing more bad numbers and, eventually, chaos: the opposite of the pairs, trios, and pretty numbers you detest.