Well erosion is a huge reason for gaps. The relevant layers are just washed or blown away. The more we go back in time, the more likely such erosion happens to a particular layer.
Then, given the relative rarity of fossilization, we simply don't expect to get a fossil from every line every 100,000 years. Or even every million years. If significant changes happen in 100,000 years and none of that small population gets fossilized, there will be a gap. The result will look like a 'sudden' change. it's like you take a picture every minute of a car moving around in a city. The motion will often look very 'jerky', even though it was not in reality.
And don't forget that much of evolution happens in small populations where genetic drift is a significant factor. Since small populations have a correspondingly small chance of being fossilized, that also leads to 'gaps'.
None-the-less, we can see how thing change over 10 millions of years if we have 10 representative during that time period. As Gould pointed out, even if the change is 'gradual' from one generation to the next, it can be quite 'sudden' on time intervals of a million years, which is still a *short* time period for most of paleontology.
I'd bet most paleontologists would *love* to have a good fossil from every 10 million year period from each line they study and for a sequence going for a hundred million years. That almost never happens. But when it does, we see the gradual evolution you seem to deny happens.