Is there any data to show if coffee lake will have greater performance at equal clock speed to skylake?
Not exactly. But there is data that is close to what you want. See the top graph on page 4.
https://newsroom.intel.com/newsroom...es/11/2017/03/14-nm-technology-fact-sheet.pdf
Skylake is the red line, Coffee Lake is the orange line. That is old data, so things may be slightly different now. But, as of this spring, Intel claims that they can be:
a) 52% less power at the same performance (implied: at the same clock and core count),
b) 26% more performance at the same power level (implied: at the same core count),
or
c) Not shown, but Intel could go for something in between. A bit more speed at the same clock and a bit less power. Think about an arrow that points down and to the right on that graph.
Intel separately claimed about a 10% performance boost on spec int. I assume that Intel did not mention floating point performance since it was less than a 10% gain. Thus, I assume that Intel is going for an arrow on that graph that goes mostly down in power and only slightly more in performance. Doing that would give Coffee Lake roughly a 10% performance boost at about 70% of the power of Skylake. Then Coffee Lake could have the same clock speeds as Skylake and 50% more cores all within about the same power envelope since (91 W for Skylake 6700K) * (1.50 for more cores) * (70% for less power) = (95 W for Coffee Lake 8600K).
Of course, Intel could have gone for a slightly different optimization too. That is just speculation.