Destination Trax™ — The Full Record

The arithmetic behind “Retire in the Wrong Decade”  ·  Kenneth Gulliver

The underlying strategy has been used with clients since 2004 and refined continuously since, and the software that manages it has been running live cases for fifteen years. Every figure on this page is that same software pointed backwards. Everything here is the working behind the article: seven market eras year by year, the contracts and caps used, every charge deducted, and the limits of what a backtest can tell you. If you want to check our arithmetic rather than take our word for it, this is the page for that.

Nobody Gets the Average

The long-run average is a story about everybody. Your retirement is a story about you.

You do not get seventy-five years of market history; you get one working career and one retirement, and which years land inside them was decided by the year you were born.

Raw S&P 500 price index, before any fee. Every S&P figure elsewhere on this page is net of 1.50% a year, which is why the same era reads lower there — 1998–2022 compounds at 5.66% raw and 4.07% after the fee.

The investor who worked from 1966 to 1982 did everything right and earned 2.50% a year while consumer prices nearly tripled. His money grew about 52% across those seventeen years; the cost of living grew roughly 198%.

He finished with close to half the buying power he started with — and his statement showed a gain the whole way down.

Two kinds of average, and only one of them spends

The simple average adds the yearly returns and divides. The CAGR is the steady rate that actually turned your starting balance into your ending balance. They are never the same when there is volatility, and the gap always runs against you. Up fifty percent then down fifty percent averages zero — but a hundred dollars becomes a hundred and fifty, then seventy-five. Both appear in every table here.

The Seven Eras

Every run starts with the same $250,000, and every figure is walk-away money — what you could have cashed out and left with, after the surrender charge, after the bonus recapture, after every fee.

Two things about how the table was built. We handicapped ourselves: every S&P figure is net of 1.50% a year, roughly what a managed account costs, because running a strategy that carries real charges against an index that carries none is not a comparison. And the two forty-year rows come from the same place as the rest: our own rules cap a contract at five years before it is rebuilt, so forty years needs eight contracts in sequence, and the DX Illustrator carries eight. Each row has been reconciled to an independent implementation of the same engine, to the cent.

Each era runs at its real length, so the ending-value columns cannot be compared down the table. Compare across, within each row.

Simple average is the mean of the annual changes in each column’s own value, so both columns are measured the same way and both compound to the ending value shown. Arizona filings, $250,000 single premium, client 54. Contracts held three to five years and then exchanged. All surrender charges, bonus recapture and product fees deducted. Benchmark is the S&P 500 price index net of 1.50% a year. Hypothetical backtest — no client achieved these results.

How It Gains in a Year the Market Falls

Inside a fixed indexed annuity your money is not in the market. You do not own the index and you do not own the shares. The carrier watches what the index did and credits interest based on it, limited by a cap, with a floor of zero. Your balance is a number the carrier owes you, not a pile of shares whose price can fall.

That floor is worth almost nothing on an average and a great deal in a real sequence. Run the actual years in the actual order and, across 1998 to 2022, the index finished lower than it started in four separate years — 2000, 2001, 2002 and 2008 — and finished a fifth, 2011, exactly flat. In every one of those five the contract credited zero.

Change in total strategy value measured from the start of that year to the start of the next, as a percentage, on the 1998–2022 Arizona illustration. The 2002 figure includes a scheduled contract exchange, which paid a new day-one bonus on the whole balance.

In not one of those five years did the balance fall.

In four of them it rose — including 2002, when the index lost 23.4% and the strategy gained 6.44%. In 2008 the index lost more than a third of its value and the strategy stood exactly still, the only one of the five that did not move at all. Standing still in 2008 is the whole product. Anyone telling you a capped strategy gains in every falling year is selling you something.

The Same Test, With a Retiree Actually Drawing Income

Everything above compares ending values without anyone taking money out. That is the wrong test for the problem this started with, so here is the right one.

Two retirees, each with $250,000, each drawing $12,500 a year — five percent of the starting balance, taken every year for twenty years. Both live through the same twenty years of market returns, at an identical average of 5.60% a year. The only difference is the order. One retires in 2000 and gets the lost decade first. The other retires in 2010 and gets the bull market first, then the lost decade.

S&P 500 price returns for 2000–2019 in both orders, benchmark net of 1.50% a year, income taken at the start of each year. Destination Trax™ run on the same twenty years with four five-year cycles, all surrender charges, bonus recapture and product fees deducted. Hypothetical backtest — no client achieved these results.

Same twenty years. Same average return. Same withdrawals. The retiree who got the bad years first ran out of money in 2016; the one who got them last finished with $155,515. Nothing separated them but the order.

Now read the two Destination Trax™ rows. The bad sequence ended at $421,503 and the good sequence at $451,085 — a difference of about seven percent, against the difference between solvent and broke.

That is what the floor is actually for. Not a bigger number — a number that stops depending on when you happened to retire.

What this test also exposes

Destination Trax™ paid $200,000 of income over the twenty years, not the full $250,000. The reason is our own rule: nothing moves in the first twelve months of any contract, so in the opening year of each of the four cycles there is no penalty-free allowance to draw from. Four years, $12,500 each.

A retiree running this strategy needs roughly one year of income in cash at each contract change — four years of it across a twenty-year plan. That is a real planning requirement, and it is the reason we tell clients to hold reserves outside the contracts.

The alternative is to draw more than the penalty-free amount in those years and pay the surrender charge, which is worse. The buffer is the right answer and it belongs in the plan from the start.

The Floor Is Standard. The Crediting Architecture Is Not.

Every fixed indexed annuity has the floor. What separates a strong contract from a weak one is the crediting architecture behind it, and the difference is enormous. An independent screen ranked 375 eligible crediting methods by what they actually returned on a $100,000 deposit over the ten years from March 2015 to March 2025, with surrender terms included. Here is the top of that list.

Top 50 of 375 eligible crediting methods, ranked by return on a $100,000 deposit over the ten years from March 2015 to March 2025, with surrender terms included. Source: Indexalyzer screening report run 24 March 2025. Some entries use partial hypothetical back-testing, as marked in the source. Rates are set by the carrier and change without warning; a screen run today would rank differently.

Four contracts out of 375 reached double digits.

The fifth managed 9.09%. The sixth fell to 7.46% — a drop of more than a point and a half between two adjacent places on the list. By the fiftieth the return was 5.34%, and 325 contracts finished below the fiftieth.

Roughly one contract in a hundred was excellent and a handful more were respectable. The rest delivered something closer to a savings rate over that decade.

The floor worked in every one of them. Not losing money is not the same as making any.

Look at the spread rather than any single line. The best contract on that list returned 12.81% a year. The fiftieth returned 5.34%. Those are the same ten years, the same market conditions and the same 0% floor — more than seven percentage points a year apart. The entire difference was which contract the buyer happened to be sold.

That distribution is why this exists as a strategy rather than a product recommendation.

The current library holds 983 crediting options read off thirty-seven carrier rate cards, and only 495 survive the screen — the rest run on multi-year terms, or on volatility-controlled and proprietary indices that do not track the S&P 500 a buyer thinks they are getting, or carry a floor below zero. From what remains, the strategy rebuilds into a new contract every few years, which means the selection has to be right not once but every single time.

Buying a fixed indexed annuity is easy. Buying the right one, four or eight times in sequence, as caps move and carriers re-file, is the part that took twenty-one years of running it with clients to refine.

What It Costs, in Full

The charges are real, they are lumpy, and they land at each contract exchange. Here is the entire 1998–2022 ledger.

Total charges $486,067 across four contract exchanges in twenty-five years, which is an average drag of 1.46 points of compound return a year — a gross 7.92% reduced to the 6.46% shown in the record above.

Every exchange pays a surrender charge and hands back part of the bonus. The case for the structure was never that it is cheap. It is that after paying all of it you finished with $1,195,856 against the index’s $677,835, and never had a dollar in the market’s path.

“Is My Money Locked Up?”

On these contracts about ten percent of the balance is available each year with no charge, and this strategy does not leave it sitting there — it takes it every year on purpose, which is the entire engine. Beyond that, a surrender schedule that shrinks each year until it disappears.

In the first year or two, walk-away value sits below what you put in, because that schedule is at its steepest and the bonus has not yet vested. On the 1998–2022 illustration the day-one walk-away figure is $236,295 against $250,000 paid in. We show that number rather than hide it.

One rule we hold ourselves to that no carrier requires: nothing moves in the first twelve months of any contract.

Several of these contracts would permit a first-year withdrawal. That a carrier allows something is not a reason to do it.

The Contracts and the Caps

The current product library holds 983 separate crediting options read off thirty-seven carrier rate cards, and only 495 of them can even be used here — the rest run on multi-year terms, or on volatility-controlled indices that do not track the S&P 500 you think you are buying, or carry a floor below zero, which would break the one promise the structure makes.

Caps and bonuses transcribed from current third-party product rate cards, not from carrier filings, on the dates shown. Verify every figure against the carrier’s current disclosure for your state before relying on any of it.

Notice the shape of that table, because it is the trade the whole industry runs on.

A carrier paying a large bonus funds it with a lower cap and a longer surrender schedule.

You are not being shortchanged — you are being handed the money up front instead of over time. Which is worth more to you is the arithmetic the whole strategy is built on.

Caps are non-guaranteed. They are declared for one contract year at a time and may reset at each renewal, subject only to the contractual minimum.

This illustration holds every rate constant for the whole period, which no real contract does. If caps fall from here, these results fall with them.

That is the single largest source of uncertainty in every figure on this page, and no historical test can measure it, because it is a future carrier decision rather than a market outcome.

The same contract is not the same contract in every state.

Caps, bonuses and surrender schedules are filed state by state. Every figure here was run on Arizona filings. Yours should be run on yours.

Zero Market Risk Is Not Zero Risk

Across all seven eras, in every single year, none of your principal was exposed to a falling index.

Zero is structural, not lucky — it is what a 0% floor means, and it held in the two eras the strategy lost.

That word does exactly as much work as it has earned and no more. It means one specific thing: no part of your principal is exposed to a falling index. It does not make the money liquid, it does not remove the insurer’s ability to pay, it does not stop caps being lowered, and a capped return can still trail inflation.

Destination Trax™ content v51 · 24 August 2026 · figures generated from DX Illustrator v73, market data through 2025.