Anti-Dilution Provision: Weighted Average vs. Full Ratchet

The clause that decides who absorbs a down round, with the math worked through

Deal Terms10 min read
Anti-Dilution Provision: Weighted Average vs. Full Ratchet

An anti-dilution provision is a term in a startup's charter that protects preferred investors if the company later sells shares at a lower price than they paid. It lowers their conversion price, so each preferred share converts into more common. Broad-based weighted average makes a modest adjustment; full ratchet reprices the whole stake and, per Cooley GO, is far less common in US venture deals.

Definition: An anti-dilution provision is a price-based protection for preferred stockholders that lowers their conversion price when the company issues stock below the price they paid (a down round), increasing the number of common shares each preferred share converts into.

Most of the time, nobody reads this clause. Then the company raises at a lower price, and it decides who pays.

Illustrative worked example: A Series A investor pays $1.00 per share for 2,000,000 shares ($2M). The company has 10,000,000 shares counted on a fully diluted basis (common, preferred as converted, and outstanding options). Two years later it raises $2M at $0.50 per share, issuing 4,000,000 new shares. Under broad-based weighted average protection, the Series A conversion price drops to about $0.857, so the 2,000,000 preferred shares convert into about 2,333,000 common instead of 2,000,000. Under full ratchet, the conversion price drops to $0.50 and the same preferred converts into 4,000,000 common.

Same round, same money. One version hands the Series A about 333,000 extra shares. The other hands it 2,000,000.

What an anti-dilution provision protects against

Investors accept percentage dilution in an up round, because their smaller slice is worth more. Anti-dilution provisions deal with one situation only: a new issuance at a lower price per share than the protected investor paid.

The protection lives in the certificate of incorporation and is summarized in the term sheet. It runs on a few supporting terms (conversion price, fully diluted shares, exempted securities), and the venture capital terms glossary defines each of them. The weighted average formula below is the one McCarter & English quotes from the NVCA term sheet, and the NVCA's model certificate of incorporation was last updated in October 2025.

How often does it actually bite? Cooley's Q2 2026 venture financing report found that 12.1 percent of its 166 deals were down rounds, and Carta's State of Private Markets report put the Q1 2026 rate at 11.4 percent. That is roughly one priced round in eight.

How the broad-based weighted average formula works

Weighted average asks two questions: how far did the price fall, and how much cheap stock was sold relative to the company's size? The formula, as McCarter & English sets it out from the NVCA term sheet, is:

CP2 = CP1 x (A + B) / (A + C)

Where:

  • CP1 is the conversion price immediately before the new issuance, and CP2 the conversion price immediately after it.
  • A is the number of common shares outstanding immediately before the new issuance, treating as outstanding all common issuable on exercise of outstanding options and on conversion of outstanding convertible securities, including preferred stock. Note that this counts options already granted, not unissued shares still sitting in the pool; some companies negotiate to add those.
  • B is the aggregate consideration received for the new shares divided by CP1 (the shares the new money would have bought at the old price).
  • C is the number of new shares actually issued.

In the example: A = 10,000,000, B = $2,000,000 / $1.00 = 2,000,000, C = 4,000,000. CP2 = $1.00 x 12,000,000 / 14,000,000 = $0.857, so each preferred share converts into about 1.167 common shares, roughly 333,000 extra shares in total.

Narrow-based weighted average uses the same formula but leaves some securities, typically options and other derivative securities, out of A. A smaller A means a bigger adjustment: if A were 8,000,000 in the example, CP2 would fall to about $0.833 and the Series A would convert into 2,400,000 shares.

So the fight over "broad" versus "narrow" is really a fight over one letter. Watch how A is defined.

Full ratchet: rare, and why founders resist it

Full ratchet doesn't care how many shares were sold. If any new stock is issued below the investor's price, the conversion price resets to that price. An investor who paid $10 per share for 1,000 preferred shares converts into 2,000 common shares if a later round is priced at $5.

Cooley GO calls full ratchet far less common in US venture deals than broad-based weighted average. McCarter & English describes it as extremely favorable to investors, because the adjustment isn't tied to the actual dilution. A tiny bridge at a lower price could reprice an entire Series A, which is why founders push back.

Anti-dilution provision math: what each version does to ownership

Here is post-round ownership in the worked example (8,000,000 other existing shares, 2,000,000 Series A, 4,000,000 new shares), as converted:

Protection Series A converts into Series A ownership Other existing holders
None 2,000,000 14.3% 57.1%
Broad-based weighted average about 2,333,000 16.3% 55.8%
Narrow-based (A = 8,000,000) 2,400,000 16.7% 55.6%
Full ratchet 4,000,000 25.0% 50.0%

The adjustment also trims the new investor, from 28.6 percent to 27.9 percent under broad-based and to 25.0 percent under full ratchet, unless their share count is set after the adjustment. That's one reason a new lead may care about the clause.

Carve-outs: issuances that don't trigger the clause

The charter lists "exempted securities" that can be issued without any adjustment. McCarter & English names the standard ones:

  1. Shares issued on conversion of convertible securities, including preferred.
  2. Stock splits, dividends, and similar changes to the common stock.
  3. Equity incentives issued from the option pool.
  4. Shares issued in certain approved transactions, such as acquisitions or strategic deals.

McCarter & English also notes it is common, and generally good for the company, to let a set percentage of investors waive the adjustment for all holders. Because protected investors can waive it, a real down round usually comes with a negotiation about whether they will.

How investors should weigh an anti-dilution provision in a live deal

From the investor's chair, the clause is insurance, and founders, employees and the next investor pay the premium. Our read on the trade-offs:

  • Broad-based weighted average is usually enough. Asking for full ratchet on a seed or Series A deal can signal that you optimize for your downside at everyone else's expense. We'd think twice before asking. Some investors see it as reasonable in distressed or high-risk rounds, and that's a fair argument in those settings.
  • Expect waiver talks. Cooley GO notes founders regularly negotiate to waive or reduce adjustments. A key question is whether management will still be motivated by its post-financing stake.
  • Check for pay-to-play. A pay-to-play provision can convert non-participating holders out of preferred, taking this protection with it. Cooley's data showed pay-to-play in 8.4 percent of Q2 2026 deals, up from about 7 percent in Q1 2026 and 6.3 percent in Q4 2025.
  • Model it before you sign. It can help to run the formula on the cap table at two or three hypothetical prices. We lean on the cap table more than the deck for a reason, and our take on reading it first explains why.
  • Remember what it doesn't cover. Anti-dilution protects against price, not percentage dilution from an up round, and it is separate from liquidation preference.

The down round guide covers how this clause interacts with new terms, the option pool, and the 409A. If you're on the other side of the table, the founder-side guide to term sheets is the better starting point.

New angels who want to watch these clauses get negotiated on live companies can join 1752vc's Emerging Angels program, which gives accredited investors a seat in a working fund's investment process for eight weeks, with live diligence calls, deal reviews, and monthly Investment Circles.

The bottom line

Broad-based weighted average spreads a down round's cost in proportion to what actually happened. Full ratchet pushes most of it onto whoever didn't negotiate for it. In our view, the first is a reasonable ask and the second is worth resisting unless the round is truly distressed.

Weighted average shares the bruise. Full ratchet hands it to someone else.

Key takeaways

  • An anti-dilution provision lowers a preferred investor's conversion price after a down round, so the investor receives more common shares on conversion.
  • The weighted average formula is CP2 = CP1 x (A + B) / (A + C); broad-based versions count outstanding options and convertibles in A, which keeps the adjustment moderate.
  • Full ratchet resets the price fully regardless of round size and is far less common, according to Cooley GO.
  • Down rounds ran at 11.4 percent of rounds in Carta's Q1 2026 data and 12.1 percent of Cooley's Q2 2026 deals, so the clause is live in about one round in eight.
  • Before signing, it is worth checking the formula, the definition of A, the exempted securities, the waiver threshold, and any pay-to-play clause.

Frequently asked questions

An anti-dilution provision is a charter term that protects preferred stockholders if the company later issues shares at a lower price than they paid. It lowers their conversion price so each preferred share converts into more common shares, shifting part of the down round's cost to common holders and to investors without the protection.

Weighted average adjusts the conversion price according to how many cheaper shares were sold relative to the company's share count, so a small down round produces a small adjustment. Full ratchet resets the conversion price to the new, lower price no matter how few shares were sold. Cooley GO describes full ratchet as far less common in US venture deals.

Under the NVCA formula, CP2 = CP1 x (A + B) / (A + C). A is the fully diluted share count before the round (including outstanding options and convertibles), B is the new money divided by the old conversion price, and C is the shares actually issued. Divide the original purchase price by CP2 to get the new number of common shares per preferred share.

Standard carve-outs cover shares issued on conversion of preferred or other convertibles, stock splits and dividends, grants from the option pool, and certain approved strategic or acquisition deals. Charters also commonly let a set percentage of preferred holders waive an adjustment for everyone, which is how many down rounds get done.

Not in the same way. A SAFE converts at its valuation cap or discount in the next priced round, which gives some price protection at conversion, but SAFE holders do not hold charter-based anti-dilution rights until they become preferred stockholders.

Sources

Disclaimer: This guide is for general education only and is not legal, tax or investment advice. Laws, market data and program terms change, so it may not reflect the latest developments or fit your situation. Treat it as a starting point, not a source of truth, and talk to a qualified lawyer, accountant or financial adviser before you make decisions.