When the Yen Surges Two Days Straight: A Decision Tree for Getting Out Cleanly

Think of what follows as a flowchart written in sentences. The yen has closed higher two sessions in a row, your platform is blinking, and someone in a Telegram group is telling you to add. I am going to ask you three questions in order. Each one narrows what a sensible exit looks like for your specific book — not the model book the InvestingLive wrap is written for. The math we will lean on is spread cost, because the post-2001 shift from 5-pip markups to raw-plus-commission changed which exits are cheap and which ones quietly bleed you. Answer honestly. The table at the end does the arithmetic.

Question 1: Is Your Position Long JPY or Short JPY Going Into the Second Day?

This is the fork that determines whether the second-day surge is running with you or against you, and it matters more than most exit guides admit. A wrap headline describing "the yen higher for a second straight day" reads the same to both sides of the book, but the exit geometry is not symmetrical. The trader long JPY — short USD/JPY, short EUR/JPY, short GBP/JPY — is watching profits accumulate and is being tempted, quietly, by the drug that has ruined more accounts than leverage ever did, which is the belief that a two-day trend is a three-day trend. The trader short JPY is watching an underwater screen, doing the mental math on whether to add, average, or fold. Same headline. Opposite psychology. Opposite exit rules.

Let me tell you what nobody in the Telegram group will say. The concession first, because the argument on the other side deserves it: yes, momentum in JPY pairs after two consecutive daily closes in the same direction has, historically, more follow-through than momentum in most G10 crosses. That is real. Yen rallies driven by risk-off tend to have a compounding quality because the currency is a global funding leg — when it moves, it moves because carry books everywhere are unwinding, and carry books do not unwind in one session.

Now the teardown. That statistical edge belongs to a book that was already positioned before day one. If you are reading a wrap on day two and deciding what to do, you are not sitting on the edge. You are sitting on the tail of it. The people who caught the move are already thinking about where to trim. Your entry, if you take it here, is the exit for someone with a lower cost basis.

If Yes — You Are Long JPY

You are in profit. The temptation is to run it. The temptation is wrong most of the time, and the reason is not psychological — it is structural. On the second consecutive day of a yen rally, one-way flow attracts intervention chatter, and intervention chatter attracts two-way price action. The volatility that just paid you can un-pay you in a single hour of Tokyo trading. Trim a third. Move your stop to breakeven on the rest. Do not add.

If No — You Are Short JPY

You are underwater and being told by a text-message chorus to double down. Do not. The exit rule here is arithmetic, not opinion. Calculate what a further 1.5% adverse move does to your account equity. If that number is more than 20% drawdown, you cut half the position at market on the London open of day three, when spreads reset from the Tokyo overnight widening. If it is less than 20%, you may hold but you do not add. The trader who averages down in the middle of a two-day trend move is the trader who becomes a case study.

Question 2: Are You Paying Raw-Spread Commission or a Marked-Up Spread?

This is the question that decides whether your exit is cheap or expensive, and almost nobody asks it on the way out — they only ask it on the way in. Which is the exact wrong order. The bid-ask spread on JPY crosses widens during any move that qualifies for a wrap-headline mention. On a normal session, EUR/JPY on a raw-spread account might quote 0.2 pips. On the second day of a yen rally with Tokyo intervention rumors circulating, the same pair on the same account can quote 1.8 pips at 3 AM London time. On a marked-up standard account, add another 1.5 to 2 pips of broker markup on top. That is a 3.5-pip spread on your exit — on a pair you might have entered at 0.9 pips.

The historical context matters here. Pre-2001, retail forex was a manual market and spreads of 5 to 10 pips on major pairs were normal because the market maker was, quite literally, a person at a desk hedging by phone. Post-2001, the electronic communication network model — brokers like FXCM Active Trader and later IC Markets Raw and Pepperstone Razor and Tickmill Pro built their businesses on it — separated the spread from the fee. The raw spread became the market's spread, quoted from a pool of liquidity providers, and the broker charged a commission on top, typically $3 to $7 per lot per side. Marked-up standard accounts still exist, but they carry the fingerprints of the old model: a wider spread that rolls both the liquidity cost and the broker's take into one number.

Here is the primary-document contradiction you need to unwind, because two things that are simultaneously true will decide your exit. IC Markets' own published commission schedule states $3.50 per side per standard lot on the Raw account, with the spread on major pairs quoted from the underlying LP pool. Their own published fact sheet for the Standard account, meanwhile, shows an average EUR/USD spread of 1.0 pip with no commission. Both are operative offers from the same broker. Both are marketed as competitive. The question the second-day yen rally forces you to ask is: which one hurts more on the exit?

If Yes — You Are on Raw-Spread Commission

Your total cost to close is spread plus commission, and during a rally-day widening the spread portion swells but the commission is fixed. On a 1-lot EUR/JPY position closed during a 1.8-pip spread window with $3.50 commission, your cost to exit is roughly $18 in spread plus $3.50 in commission — call it $21.50 all-in. That is expensive relative to a normal session but cheap relative to your alternative, because on the raw account the spread widens with the market, not with the broker's judgment. You exit at market. You take the cost. You do not wait for the spread to "come back" because during a two-day rally there is no coming back until Tokyo closes on day three.

If No — You Are on a Marked-Up Standard Spread

Your total cost to close on the same 1-lot EUR/JPY position during the same rally window is roughly 3.5 pips, or $35 all-in, and the marked-up portion is opaque — you cannot tell how much is market and how much is broker. This changes your exit geometry. You do not exit at market during the widest part of the Tokyo session. You wait for the London open, which typically compresses spreads back toward the daily average, and you exit into that compression. If your platform lets you set a limit order to close at a specific price rather than market, use it. On a marked-up spread, the market order is the tax.

Question 3: Do You Have a Written Exit Rule Older Than This Move?

This is the question that separates traders from tourists, and the answer is either yes or no with no defensible middle ground. A written exit rule older than the current move is a rule that was decided before you had emotional exposure to the outcome — before the account balance was flashing, before the Telegram messages were arriving, before your girlfriend asked how it was going. If you wrote your exit rule during the rally, it is not a rule. It is a rationalization with a rule-shaped costume.

I know how this sounds. Let me give the concession its due: some of the best exits in the archival trader-memoir record were improvised. There are documented cases in published trading literature from the 1990s onward of position closures decided in the moment based on observed market microstructure — a sudden thinning of the bid, a rejection of a specific price level, a change in the character of the flow. Improvised exits are not always wrong. Traders who read the tape well can adjust in real time.

Now the teardown. Those improvised exits were made by people who had already made ten thousand trades and had, by any reasonable measure, internalized their exit framework so deeply that the improvisation was not really improvisation — it was pattern recognition executing faster than they could articulate. If you have been trading yen crosses for six months and you decide, at 4 AM watching a wrap headline, that today is the day to freelance your exit rule because "this move is different," you are not improvising. You are hoping. Hope is not a rule.

If Yes — You Have a Pre-Existing Written Rule

Execute the rule. Do not second-guess it. Do not adjust it because the wrap headline used the word "surges" instead of "rises." The rule was written by a version of you who was not in this position and did not have this specific dopamine dependency. That version of you was smarter than the current version of you, at least for the purposes of this decision. If the rule says exit at a 2% adverse move from entry, exit at a 2% adverse move from entry. If it says exit half at a 3% favorable move, exit half at a 3% favorable move. The rule wins.

If No — You Have No Pre-Existing Rule

You are now writing the rule under duress, which is exactly the scenario the rule was supposed to prevent. Accept that this trade is compromised at the exit level and do the minimum-viable version: close half the position immediately at market, set a trailing stop 1% behind current price on the remainder, and — this is the part you will resist — write down the rule you should have had before the next trade. Save it. Timestamp it. It becomes the pre-existing rule for the next situation.

If You Answered Everything: The Nine-Row Decision Table

The three questions above generate more than eight combinations because Question 3 has a "yes" branch with two sub-cases (rule says hold vs rule says exit), but for the workbook version we collapse those into a single "follow the rule" recommendation. Read your row.

One paragraph of context on the ninth row, because it will bother the arithmetic-minded reader. The ninth row overrides the other eight when equity risk crosses a specific threshold. Spread cost matters until it does not, and once your remaining equity is threatened by a further ordinary-sized move, the cost of the exit is trivial next to the cost of not exiting. This is the row that separates traders who survive their first bad month from traders who do not.

What This Piece Does Not Cover

This decision tree does not address the tax treatment of realized JPY gains or losses in your jurisdiction, which is not a small caveat — a trader in Singapore, a trader in India using an offshore broker, and a trader under UK spread-betting rules face genuinely different arithmetic on the same closed trade, and the exit that is optimal pre-tax is not always optimal post-tax. It does not address hedging strategies that use options to reshape the exit rather than close the underlying, because AvaTrade's AvaOptions platform and similar venues make partial-hedge exits genuinely viable for some books but require a separate analytical framework that this piece cannot compress. It does not address the intervention question — whether the Bank of Japan's Ministry of Finance directive apparatus is or is not active on the specific dates you are trading — because that requires reading the current statements out of Tokyo and is a moving target that a written decision tree cannot capture without going stale within a week. Each of those is its own piece, and each deserves the full grounding treatment rather than a paragraph tacked onto this one.

FAQ

Does the "second consecutive day" pattern in JPY actually have a statistical edge?

There is documented follow-through in yen rallies driven by risk-off unwinds because the currency serves as a global funding leg, and carry-book unwinds do not resolve in a single session. However, the edge belongs to traders positioned before day one — reading the wrap on day two puts you on the tail of the move, not the front. Treat the pattern as a caution, not an entry signal.

Why do spreads widen so dramatically during a two-day yen rally?

Two forces converge. First, liquidity providers pull quotes when directional flow becomes one-sided, so the raw spread from the underlying pool widens naturally. Second, on marked-up standard accounts, brokers can layer additional markup during volatility because their pricing is not directly tethered to the LP quote. On a raw-spread account with a broker like IC Markets Raw or Pepperstone Razor, you see the market widening; on a marked-up account, you see market widening plus broker discretion.

Is exiting at market during Tokyo overnight always the wrong move?

Not always, but usually. Tokyo overnight typically shows the widest spreads of the twenty-four-hour cycle on JPY crosses, particularly during a rally week. If your position size and equity risk allow you to wait for the London open, the compression is worth waiting for — spreads on major JPY pairs often halve within the first hour of London trading. If drawdown is threatening equity, you exit at Tokyo prices and accept the tax.

How did retail spread cost actually change between 2001 and 2026?

Pre-2001, retail forex ran on a manual market-maker model with typical spreads of 5 to 10 pips on major pairs and no separate commission — the market maker's margin was baked into the quote. Between 2001 and 2010, electronic communication networks emerged, and brokers began separating the spread (routed from a liquidity provider pool) from the fee (a per-lot commission). By the 2020s, raw-spread commissions typically run $3 to $7 per lot per side, with the spread quoted at fractional pips during normal sessions.

What is the difference between a raw-spread account and a standard account in practical terms?

On a raw-spread account, you see the market's spread — often 0.0 to 0.3 pips on major pairs during liquid sessions — and pay a fixed commission on top. On a standard account, you see a marked-up spread — typically 0.7 to 1.5 pips on the same pairs — with no separate commission. Total cost is comparable on average, but raw accounts are more transparent during volatility because the widening is market-driven, not broker-discretionary.

Should I use a limit order or a market order to close a JPY position during a rally?

Market orders execute immediately at whatever bid or ask is showing, which during Tokyo widening can be punitively expensive. Limit orders let you set a specific exit price, which gives you control over the spread tax but exposes you to the price moving further away before your limit fills. On raw-spread accounts, market orders are usually acceptable. On marked-up accounts, limit orders during the London open compression window typically produce better fills.

What counts as a "written exit rule older than the move"?

A rule that existed in writing — in a trading journal, a spreadsheet, a broker's platform as a preset stop or take-profit — before the current move began. Rules invented mid-rally do not qualify because they are subject to the emotional distortion the rule framework is supposed to prevent. If you cannot point to a timestamped record of the rule predating the trade, treat yourself as being in the "no written rule" branch of Question 3.

Can I use options to exit a JPY position instead of closing the spot or CFD trade?

On a platform that supports currency options — AvaOptions on AvaTrade, for instance — you can construct a partial hedge that caps downside on an open position without fully closing it, which is useful if you have tax or structural reasons to keep the underlying trade open. The cost is the option premium, which during a two-day rally is elevated because implied volatility is bid. This piece treats options as out of scope; the analytical framework for option-based exits requires its own dedicated treatment.