Working theory on why VC has so much trouble disrupting itself and why there's so much conformity among the non-consensus crowd.
First let's take a look at a typical venture success story: Airbnb. From YC to Series F, 12 lead investors & 54 total investors https://www.crunchbase.com/organization/airbnb/funding_rounds/funding_rounds_list

VC funds are frequently set up to 1/ focus on a particular stage of funding (eg "seed") and 2/ typically lead part of a round a seek co-investors to fill out the rest.
This means companies need to get handed off to a new cohort of funds at each round adding up to dozens of firms

There's only ~1,000 VC funds total. So each round involves a non-trivial amount of all the funds active in that stage backing the company. You often hear you want to be "non-consensus and correct" but you really want to be just slightly ahead of consensus.

If the genuine consensus among Series B funds is your portfolio company is in a bad market, your company is dead (assuming their business plan and your financial model relies on them raising a Series B). So the rational way to play the game is to at some level optimize this.

The way this market is set up there are tons of incentives for eg seed funds to be very friendly with Series A funds etc. Indeed many seed funds explicitly market their ability to make warm intros and help close the next round (even tho they won't be leading it)

This has 2 bad effects 1/ VCs get further ahead drinking the kool aid and debating the inside baseball than w/ fundamental critiques. In this game, genuinely challenging other funds is actually harmful to your and your portfolio's success, hurting their chances of future funding

2/ it becomes rational to invest in fads & hype cycles. If your investment calculus includes likelihood this founder can raise the future funds they need, you end up following the trends to give your co's an edge in subsequent rounds and avoid truly non-consensus ideas

There seem to be two obvious ways around this:
1/ Have a mega fund that can back a founder from seed to IPO (a trend we're seeing but one that I think has structural advantages).
2/ Build a model that doesn't rely on follow on funding (like ours at @earnestcapital)

We are able to slaughter sacred cows like $10B markets, power laws, and growth at all costs, because we don't care what other funds think. We actually look for businesses that were tried before, got traction, and then died b/c later stage VCs wouldn't fund it.

Micro prediction (/hope): this cycle of "Access VC" (funds with strategies premised on squeezing in allocation to the hot deals) will die out soon and we'll see more strategies built on the ability to make genuinely non-consensus bets that challenge some accepted dogma.