Underwriting · 8 min read

Cash-on-Cash vs. Cap Rate vs. Total ROI — Which One Actually Matters

By Cam Burke·May 17, 2026

Three metrics, three different jobs. Use the wrong one and you'll buy the wrong deal.

Cash-on-cash. Cap rate. Total ROI. Every investor talks about all three. Half of them use the words interchangeably. That's how people end up at the closing table on a deal that pencils on paper and bleeds them dry six months in.

Here's the deal: each of these metrics answers a different question. Pick the wrong one for the deal type you're looking at and the number lies to you. Not because the math is wrong — because you're measuring the wrong thing.

I run 70+ rental units in Oklahoma City and underwrite flips, BRRRRs, and long-term rentalsevery week. This is how I actually use these three metrics — and where I've watched newer investors get hurt confusing them.

1. The Definitions (Tight and Accurate)

Before you can use these correctly you have to know what each one actually measures. Most blog posts get this part fuzzy. We're not going to.

Cash-on-Cash Return (CoC)

Annual cash flow ÷ cash invested. Measures how hard each dollar you left in the dealis working for you. Numerator is annual cash flow after debt service. Denominator is the cash you actually still have stuck in the property — down payment, closing costs, rehab cash, any money the refi didn't recycle.

Cap Rate

Annual NOI ÷ property purchase price (or current market value). Measures the property's unlevered yield — what it would return if you bought it all-cash. NOI = effective rent minus operating expenses. No debt service. Cap rate is a property metric, not a deal metric.

Total ROI (Annual ROI)

(Annual cash flow + annual principal paydown) ÷ total cash invested. Some operators tack appreciation on top — I don't, because appreciation is a guess. Cash flow plus principal paydown is math the bank will fund tomorrow.

Three different numerators. Two different denominators. They are not the same metric in different clothes. Which one to optimize for depends entirely on what kind of deal you're looking at.

2. When Cash-on-Cash Is the Right Metric

CoC is the right tool for one job: telling you whether stuck capital is doing real work.

Best use case: a partial BRRRR where the refi didn't recycle all your money. Say you closed a BRRRR, refinanced, and ended up with $8k still in the deal. That $8k could be in a HYSA earning 4.5%. It could be in another deal. Or it could be in this property generating $1,200 a year of cash flow, which is 15% CoC. That number tells you something real — your locked-up dollars are working at 15%.

Compare that to a different BRRRR with $8k stuck generating $400/yr in cash flow. Same money trapped, same risk profile, but only 5% CoC. The second one is a worse outcome even if the deal itself looked fine on the underwrite.

Where CoC misleads:on a full BRRRR where you got all your money back. If your cash invested is zero, CoC is mathematically infinite. The metric breaks. People will tell you they did a full BRRRR with "infinite return" — that's not a flex, it's a math limitation. Switch to total ROI on full BRRRRs and the actual answer comes back into focus.

Second place where CoC misleads: comparing across leverage levels. A 25% CoC on a deal with 90% leverage isn't comparable to a 15% CoC on a 60% leverage deal. The first one carries more risk — higher debt service, thinner margin for vacancy, worse downside if rates move at refi. Same number, different risk. CoC strips leverage out of the conversation, which is sometimes exactly what you want and sometimes exactly what hurts you.

3. When Cap Rate Is the Right Metric

Cap rate is the right tool for two jobs: pricing a property against the market, and comparing properties apples-to-apples.

Use case: you're looking at three rentals across different neighborhoods in your buy box. Different price points, different rents, different unit counts. Cap rate strips out how you might finance each one and tells you what intrinsic yield each property produces. A 7% cap is mechanically better than a 5% cap — assuming the risk profile (neighborhood, tenant class, deferred maintenance) is comparable.

That last clause is doing a lot of work. A 9% cap in a war zone isn't better than a 6% cap in a B-class neighborhood — it's a 9% cap because that's what the market demands to compensate for the risk. Cap rate is meaningful within a tier, not across tiers.

Where cap rate misleads:when you're using leverage. Especially right now, with debt at 7-8% on investment property, cap rate alone tells you almost nothing about whether the deal works.

Example: a 6% cap rate property with 5% debt. Positive leverage — every dollar borrowed earns more than it costs. The deal works.

Same property, same 6% cap, but with 8% debt. Negative leverage — every dollar borrowed loses money. The deal pencils worse the more you finance it. Yet a screening tool that only shows cap rate will rank these two properties identically.

The screen for this is simple: if your interest rate is higher than your cap rate, leverage hurts the deal. You either need a lower price, a lower rate, or a different deal.

4. When Total ROI Is the Right Metric

Total ROI is the right tool for long-term rental decisions where you're holding seven-plus years.

The reason is principal paydown. On any amortizing mortgage, your tenant is paying down the loan a little bit every month. Year one of a 30-year mortgage on $150k at 7%, you're paying down about $1,500 of principal. Boring. By year ten you're paying down closer to $3,000 a year, and total principal paydown across the decade lands around $25k. That's real money on a deal where your starting cash was maybe $35k.

CoC ignores all of that. It only counts cash flow. So a rental with a mediocre 6% CoC looks pretty average — until you remember it's also building $2k-$3k of equity per year on autopilot. Total ROI captures that. CoC doesn't.

On a long hold, total ROI is closer to the truth of what the deal returns.

Where total ROI misleads:on short holds. If you're flipping, or doing a quick BRRRR you'll sell inside two years, principal paydown is so small it doesn't matter. Focus on cash returns and net profit.

Second place it misleads: when operators inflate it with assumed appreciation. "My total return is 22% — 8% cash, 6% principal, 8% appreciation." That last 8% is a guess. Cash flow is real. Principal paydown is contractually defined. Appreciation is wishful thinking until the day you sell or refinance and a third party confirms it. I keep appreciation out of total ROI for that reason. If the deal needs appreciation to pencil, it's a speculation, not an investment.

Run your deal with all three metrics plus DSCR in one pass.

Open the Rental Property Calculator

5. Which Metric for Which Deal Type

This is the chart in my head every time I open a new deal. Match the metric to the strategy.

Fix-and-flip

None of these three. Measure net profit and total ROI on cash invested over the project life (typically 4 to 9 months). Annualizing a 6-month return is misleading — you can't actually run that deal twice in a year reliably. Stick to dollars-in, dollars-out, project ROI.

BRRRR with all money out (full BRRRR)

Cap rate to size the asset against the market. Monthly cash flow as the operating signal. DSCR to confirm the refi works. CoC is broken (zero denominator). Total ROI on your starting cash is the meaningful return.

BRRRR with money stuck (partial BRRRR)

CoC matters intensely here — your stuck capital needs to earn its keep. 8%+ minimum on the trapped dollars. Cap rate still useful for benchmarking. DSCR for the next refi or sale option.

Long-term rental, 7+ year hold

Total ROI is the truest number. Monthly cash flow is the operational signal — has to be positive every month or you're subsidizing the tenant. Cap rate for comparison shopping.

Short-hold rental (2-3 years before sale or refi)

CoC and monthly cash flow. Don't lean on total ROI because principal paydown over 24-36 months is small enough to ignore. Focus on cash yield and exit value.

6. The Three Misuses I See Constantly

These are the specific mistakes that put newer investors in bad deals. None of them are exotic. All three show up in deal pitches every month.

Misuse 1: Comparing CoC across different leverage levels

"Deal A is 25% CoC, Deal B is 15% CoC, so Deal A is better." Not if Deal A is 90% leverage and Deal B is 60% leverage. Deal A has thinner margin for vacancy, more sensitivity to rate moves at refi, and a worse downside if rents come in soft. Same number, different risk profile. CoC has to be compared inside a similar leverage tier or it lies to you.

Misuse 2: Using cap rate as a verdict

"It's a 7% cap, so it's a good deal." Maybe. Depends entirely on what you're paying for the money. A 7% cap with 6% debt is positive leverage and works. A 7% cap with 8% debt is negative leverage and loses money the more you borrow. Cap rate alone isn't a verdict — it's a property descriptor. Pair it with your interest rate to know if leverage helps or hurts.

Misuse 3: Total ROI inflated with appreciation

"My total return is 22%." Sounds great. Then you find out 8 points of it are assumed appreciation. Appreciation is a guess until the property sells or appraises. Cash flow and principal paydown are math. Don't conflate them. I underwrite with zero appreciation assumption — anything I get on top is upside, not the thesis. The deal has to work without it.

7. DSCR — The Metric That Decides Whether the Deal Closes

One more metric, then we're done. This one isn't a return metric — it's a gate.

DSCR (debt service coverage ratio) = NOI ÷ debt service. Banks underwriting investment property refis require a minimum DSCR — typically 1.20 to 1.25 on standard product, sometimes 1.0 on aggressive DSCR-loan programs.

If your DSCR is 1.25, that means for every $1.00 of debt service, the property generates $1.25 of NOI. The bank wants that 25% cushion in case rents soften or expenses spike.

Here's why it matters: if your deal doesn't hit DSCR, none of the other metrics matter. You can't get the loan. The 15% CoC, the 7% cap, the 18% total ROI — all irrelevant if the lender won't fund the refi. The deal sits in hard money longer, costs you more, and either dies or forces a fire sale.

Add DSCR to your underwriting checklist as a pass/fail gate. Run it before you run anything else. If a deal fails DSCR at the lender's minimum, fix the deal (lower price, higher rent, longer amortization) before you spend another five minutes on return math.

The Short Version

Three metrics, three jobs. Pick the right one for the deal you're looking at.

CoC — how hard your stuck capital is working. Best for partial BRRRRs and short-hold rentals.
Cap rate — the unlevered yield of the property. Best for benchmarking against the market. Useless without checking it against your interest rate.
Total ROI — cash flow plus principal paydown over total starting cash. Best for long-term holds where principal paydown compounds.
DSCR — the lender's gate. Run this first or none of the above matters.

And the one rule that runs underneath all of it: never let one metric carry the verdict.A great CoC with a terrible cap rate means you bought too high and leverage saved you. A great cap rate with a failing DSCR means you can't close. A great total ROI driven by assumed appreciation means you're speculating.

The deal either works on the three real metrics together — cash flow, cap rate, and total ROI — and passes the DSCR gate, or it doesn't work. Stop letting a single big number talk you into a deal the other numbers are quietly telling you to walk from.

Underwrite the whole deal, not one number.

VAC runs CoC, cap rate, total ROI, DSCR, and monthly cash flow on every deal — flip, BRRRR, or long-term rental.

Run a Rental Deal Start 7-Day Free Trial

Keep Reading

The 75% Rule Lies

A BRRRR at 73% LTV that still loses — and why the rule is a screen, not a verdict.

Multi-Unit BRRRR Underwriting

How the math shifts moving from 1 unit to 4 — vacancy variance, expense allocation, DSCR.

Hard Money Cost on a 6-Month BRRRR

The 4-10 grand most calculators skip — what real hard money math looks like.