A daily pivot line is a calculated reference, not an instruction to buy or sell on contact. This strategy tests a narrower idea: after price rejects S1 from above or R1 from below, does a separate completed M15 confirmation leave enough executable room toward the central pivot? The rules below specify what must be known, when an attempt expires and how costs can reject an attractive chart. All prices, paths and ledger counts are fictional teaching examples. No profitable track record is claimed.
1. What this pivot-bounce hypothesis actually tests
The proposed effect is a return from a first support or resistance reference toward the central pivot after observable rejection and follow-through. It is a mean-reversion hypothesis with a price-confirmation gate. A touch alone, an attractive wick alone and a later return to P are insufficient. The same level can break decisively; the strategy accepts ordinary losing trades rather than redefining every failure as a different market condition.
Use EUR/USD and M15 Bid candles as the baseline research specification. Calculate Traditional daily P, S1 and R1 from the same broker’s completed D1 Bid data. Other pairs, H1 entries, different pivot families or another session create separate specifications. The teaching parameters are deliberately explicit, not advertised as optimal settings. A reliable implementation may still produce a negative result after realistic costs.
Read the pivot formulas and broker time-zone guide first if your indicator values differ. This article develops the first strategy in the pivot cluster. A breakout-and-retest entry or a Camarilla reversal uses different conditions and should not be substituted after a stopped bounce. Educational rules do not establish that a suitable trade exists today.
For a completed break, first later retest and separate confirmation toward R2/S2, see the pivot breakout-and-retest strategy.
2. Calculate the previous-day map and retain its provenance
For the latest completed broker D1 bar, save its high H, low L and close C, start/end timestamps and data source. Use P = (H + L + C) / 3, S1 = 2P − H and R1 = 2P − L. The current D1 high, low and unfinished close never enter this calculation. Keep full calculation precision for comparisons; round submitted order prices separately to the permitted tick grid.
Our fictional previous day has H = 1.1080, L = 1.1000 and C = 1.1040. Therefore P = 1.1040, S1 = 1.1000 and R1 = 1.1080. The neat numbers are chosen to make the examples auditable, not to imply that pivots normally coincide with prior extremes. Reject missing, zero or internally inconsistent OHLC data; a valid bar requires L ≤ C ≤ H and H > L.
Use one saved broker calendar throughout the experiment. If the latest completed D1 is a short Sunday/holiday session or lacks expected data under that calendar, skip the current trading day rather than silently selecting an older bar. Do not merge Sunday data only after seeing a result. Broker-midnight and New York-close days may produce different maps, and server offsets may change with daylight saving time.
3. Freeze the clock, band and execution parameters
At 08:00 broker server time, save the pivot map and ATR(14) calculated on completed M15 candles, including the candle opened at 07:45. Freeze the half-width w = 0.2 × ATR for that day’s observation window. In the example ATR = 0.0010, so w = 0.0002, or 2 pips. The S1 band is [1.0998, 1.1002] and the R1 band [1.1078, 1.1082]. These are contact tolerances, not protective-stop distances.
The baseline permits contact bars opening at or after 08:00, requires B to close strictly before 16:00 and requests a time exit at 16:30, all in the same broker clock. These are arbitrary research hours, not a claim that every broker’s 08:00 is London’s open. Save the date-specific UTC offset. A new bar becomes usable only when a subsequent quote establishes that it has closed; nominal clock time alone does not supply missing market data.
Fix a stop/target buffer b = 1 pip, maximum entry spread s = 2 pips, adverse-fill allowance a = 1 pip, maximum quote age 2 seconds and entry deadline 5 seconds after B’s scheduled close. Require a conservative net reward-to-risk ratio of at least 1.5. For EUR/USD here, pip = 0.0001 and a five-decimal point = 0.00001. Confirm the actual tick increment; pip, point and tick are not interchangeable.
| Parameter | Baseline value | What is fixed |
|---|---|---|
| Market / candles | EUR/USD · M15 Bid | Previous completed broker D1 |
| Contact half-width | w = 0.2 × ATR(14) | Saved at 08:00; example 2 pips |
| Buffer / spread cap | b = 1 pip · s = 2 pips | Order levels and first-quote test |
| Adverse allowance / cost gate | a = 1 pip · ratio ≥ 1.5 | Before order submission |
| Quote age / deadline | 2 seconds / 5 seconds | First subsequent quote only |
| Daily limit / risk budget | 1 filled entry · 0.5% | Round volume down; no loss guarantee |
4. Treat the first contact as a consumable attempt
A contact occurs when a completed candle’s [low, high] intersects a frozen level band. Use the first such contact inside the window at each of S1 and R1. Save the preceding completed M15 close as the approach reference; at the window start this is the 07:45 candle’s close. The attempt is consumed even if the contact fails to form a qualifying rejection. Later prettier candles at that level do not restart the attempt that day.
For a BUY approach, the previous close must be strictly above S1 + w and below P. For a SELL approach, it must be strictly below R1 − w and above P. An approach already inside the band, an equality at the required boundary or a missing preceding bar fails. If one contact candle overlaps both bands, disable both attempts: OHLC cannot establish a useful order of visits. Reject a gap across the entire band: BUY if A opens below S1 − w after that approach close, SELL if A opens above R1 + w. Do not assume intervening prices traded.
Only one strategy position may be open, and at most one filled entry is allowed per broker day. Thus a failed S1 attempt may leave the first R1 attempt available, but a filled trade uses the day’s entry allowance even if it later loses. During an unresolved order request, lock further submissions until broker tickets and positions are reconciled. These limits prevent repeated attempts from quietly replacing the specified first-contact experiment.
5. BUY rules: S1 rejection A followed by confirmation B
Call the first contact candle A. A valid lower rejection requires A.low < S1 − w and A.close > S1 + w, with A.high below the long target P − b. The approach condition from the previous section must also hold. These strict inequalities distinguish a completed rejection from a close within the band. A low exactly on the lower edge or a close exactly on the upper edge does not qualify. In the notation, high, low and close mean the candle’s high, low and closing price; the letter identifies A or B.
B must be the immediately following M15 candle. Require B.close > A.high, B.low ≥ A.low and B.high < P − b. An equal high break is insufficient; an equal low is permitted because the protective stop is another b below A.low. A new lower low, a target touch before entry or failure to close beyond A.high cancels the attempt. Do not search later candles for a replacement B.
After B closes, inspect the first received Bid/Ask snapshot. Require Bid > A.high and Ask ≤ B.close + s, along with the common freshness, deadline, spread and cost checks. The planned market BUY uses Ask; its fixed stop is Bid at A.low − b and its fixed target is Bid at P − b. Tick-round the stop away from entry and the target toward entry, then recalculate distances and size before sending.
6. SELL rules: mirror the signal, state the Ask protection
For the first R1 contact A, require A.high > R1 + w and A.close < R1 − w. Its low must stay above the reference-side target boundary P + b. B is the immediately following M15 candle: B.close < A.low, B.high ≤ A.high and B.low > P + b. Equality is allowed only for the no-new-extreme condition, not for the close through A.low or the target boundary.
The next quote must retain Bid < A.low and Bid ≥ B.close − s. Enter a qualifying SELL at Bid. Freeze its protective Ask stop at A.high + b + s and its Ask profit target at P + b + s. Here s is the predeclared 2-pip allowance, not a reconstructed historical Ask high. With entry spread no larger than s, the stop includes a margin beyond the Bid rejection high. This does not guarantee that later spread widening cannot trigger it.
At a constant 2-pip spread, the short target corresponds to Bid P + b, before the central pivot. A later wider spread changes that correspondence while the submitted Ask target stays fixed. Never treat a Bid high as an observed Ask high or claim a short exit from a Bid touch alone. The market, limit and stop order guide explains the execution sides and why a resting limit at R1 would be another strategy.
7. Respect availability, the entry deadline and cancellation priority
An A candle opened at 09:00 is not a confirmed rejection at 09:00. It becomes available after 09:15. The following B opened at 09:15 becomes available after 09:30. Only then may the first new quote be assessed. A backtest that buys A’s low or B’s earlier breakout price has changed both the information set and the execution rule. All example times below are broker server times and candle labels denote opening times.
Use only that first snapshot. If it is stale, arrives more than 5 seconds after B’s scheduled close, exceeds the 2-pip spread cap, loses price confirmation or fails room/risk checks, log a skip. Do not wait for a more convenient second quote. The final permissible B closes before 16:00; a B closing exactly at 16:00 fails. Late arrival cannot be repaired by assigning an order to the nominal closing time.
Apply cancellations before entries: bad history or calendar, double-band/gap contact, invalid A, invalid or expired B, occupied/day-used state, then quote, order-price, cost and volume checks. A missing B cancels; a rejected request is logged and not automatically retried. If acknowledgement is uncertain, reconcile it before deciding whether the daily filled-entry allowance has been consumed. A later signal cannot resolve an unknown position.
8. Worked BUY: trace the decision before measuring reward
Using the saved S1 = 1.1000, P = 1.1040 and w = 0.0002 map, suppose the preceding close is 1.1008. A opened at 09:00 has open 1.1005, high 1.1006, low 1.0995 and close 1.1003. It is a first-contact lower rejection, known after 09:15. B opened at 09:15 has open 1.1003, high 1.1010, low 1.1001 and close 1.1007. Its confirmation becomes available after 09:30.
Assume the first fresh quote at 09:30:01 is Bid 1.1007 / Ask 1.1009 and the market BUY fills at 1.1009 with protection accepted. Spread is 2 pips, Bid remains above A.high, and Ask is exactly B.close + s, which is permitted. The fixed Bid stop is 1.0994 and the Bid target 1.1039. Gross risk is 15 pips and potential gross reward 30 pips, giving 2.0 before commission or additional execution differences.
The chart shows a plan after entry, not a later winning route. For a separate losing continuation, the next Bid can fall to 1.0993 and produce a stop fill at 1.0993 rather than the requested 1.0994: 16 pips of price loss. That is a possible strategy loss, not evidence that the earlier confirmation was unavailable. Stops request an exit; they do not guarantee the requested price or maximum loss.
9. Worked SELL: preserve the same rules with different quote sides
Consider a separate R1 scenario with the same fictional pivot values; it is not a second entry on the BUY example’s day. The preceding close is 1.1073. A opened at 10:00 has open 1.1075, high 1.1085, low 1.1074 and close 1.1077. B opened at 10:15 has open 1.1077, high 1.1079, low 1.1071 and close 1.1073. A is known after 10:15 and B after 10:30.
Assume the first fresh quote at 10:30:01 is Bid 1.1073 / Ask 1.1075 and the SELL fills at Bid 1.1073. Ask stop = 1.1085 + 0.0001 + 0.0002 = 1.1088. Ask target = 1.1040 + 0.0001 + 0.0002 = 1.1043. The entry-to-exit distances are again 15 pips of gross risk and 30 pips of potential gross reward. With constant 2-pip spread, corresponding Bid stop/target are 1.1086 and 1.1041.
The earlier A.high 1.1085 is not a post-entry stop event. Conversely, an Ask spike after entry may reach 1.1088 while the visible Bid remains below that number. A short stop can therefore be real even if a Bid-only chart appears not to touch its plotted Ask level. Keep tick-side data and broker records rather than resolving this by moving the stop or redrawing R1.
10. Reject insufficient room after costs and adverse-fill allowance
For each candidate, let r be the executable stop distance in pips and g the executable target distance. Convert round-trip commission to an equivalent c pips using the actual account-currency pip value. The entry gate is (g − a − c) / (r + a + c) ≥ 1.5, with positive numerator and denominator. The allowance a = 1 pip covers a single assumed adverse entry-or-exit price difference in the planning model; it is not subtracted at both ends again.
For either worked trade, r = 15, g = 30 and c = 0.7 under USD 7 commission per standard lot and USD 10 per pip. The conservative planning ratio is (30 − 1 − 0.7) / (15 + 1 + 0.7) = 28.3/16.7 ≈ 1.69. If fills match the quoted plan without that extra adverse difference, the commission-adjusted ratio is 29.3/15.7 ≈ 1.87. State which ratio is being reported.
In a separate BUY variant with B.close = 1.1009, the first snapshot Bid 1.1009 / Ask 1.1011 satisfies the 2-pip spread and overshoot limits. With the original stop and target, r = 17 and g = 28; the conservative ratio becomes 26.3/18.7 ≈ 1.41, so skip. Do not move the fixed target beyond P to rescue it. The baseline obstacle map contains prior D1 H/L and zero-anchored 100-pip round numbers only, with halves and quarters disabled. Skip if a reference lies strictly between the quote Bid and the reference-side target boundary: P − b for BUY, P + b for SELL. Save this map before 08:00; additional swing-zone filters need their own version.
11. Keep chart conditions separate from order execution
Check the symbol’s valid price grid, minimum stop distances, volume limits, available margin and trading permissions. In MQL4, MODE_TICKSIZE is expressed in points; multiply by Point to obtain its price distance. MODE_STOPLEVEL = 0 can still coexist with dynamic restrictions. Decimal display precision alone is not a tick-size test, and an accepted chart annotation is not an accepted server order.
This baseline requires a broker/execution route that accepts the protective stop and target with the opening request. If that protection cannot be accepted, skip rather than open an intentionally unprotected position. Do not widen the planned stop to overcome rejection. An unknown response must be reconciled with order tickets before another submission. Record requested prices, actual fills and server errors separately; a rejected trade is not a filled loss or win.
After a fill, recalculate actual distances and planned account-currency loss. Reduce the remaining a allowance by any realized adverse entry difference, floored at zero, instead of adding the full allowance twice. If the stop/target is absent, the 1.5 cost gate is breached or the planned loss exceeds the saved budget, request an immediate close under the recorded exception rule; retain its real cost and outcome. This is an exception response, not a guaranteed undo. If both exits appear touched inside one historical candle, use ordered Bid/Ask ticks or mark the outcome ambiguous; do not assume the profitable exit occurred first.
12. Size the position in the account currency and round down
For a USD 5000 equity account and 0.5% risk budget, the budget is USD 25. Assume EUR/USD, 100000 EUR per standard lot, USD 10 per pip per lot and USD 7 round-trip commission per lot. With 15-pip stop distance plus 1-pip adverse allowance, planned loss per lot is (15 + 1) × 10 + 7 = USD 167. Raw volume is 25/167 = 0.149700… lot.
For a 0.01 lot step, round down to 0.14 lot. Planned loss is USD 23.38; rounding up to 0.15 lot would give USD 25.05 and breach the budget. With unchanged illustrative fills, the 0.14-lot gross stop loss is USD 21, plus USD 0.98 commission: USD 21.98. Potential gross target gain is USD 42, less USD 0.98: USD 41.02. These conditional amounts are not observed results.
Verify the actual contract and tick value in the deposit currency; USD as quote currency does not imply a USD account. Account for conversion, broker minimum/maximum lots and margin separately. If the minimum permitted volume exceeds budget, skip. Slippage, gaps, spread widening and conversion changes may exceed the allowance. The position-sizing guide gives the general method; the trading-cost guide explains spread, commission and swap.
13. Use fixed exits and a finite daily allowance
Keep the submitted stop and target unchanged: no averaging down, partial exits, trailing stop, automatic break-even or reverse trade belongs to this baseline. Request a market close at 16:30 server time if a position remains. Use the actual close fill, not the 16:30 candle’s ideal price. If quotes or trading access are unavailable, record the failed time exit and continue reconciliation; being flat by the deadline cannot be guaranteed.
One filled entry exhausts the current day’s allowance. Preserve that state across terminal restarts using the saved broker day, symbol, strategy identifier and reconciled history. A request that timed out may already have filled. Reset only after confirming the new broker day and the absence of an old open or unresolved position. A forced overnight carry is an exception record, not permission to trade a fresh map alongside it.
The baseline has no discretionary news exclusion: keep scheduled-event labels in the research log. If you later add a blackout, define its calendar source, historical release timestamps and exact interval before testing a new version. A news filter chosen only for losing examples is not a valid explanation of unchanged rules. Keep any account-wide exposure limit as a separately declared pre-entry condition.
14. Backtest the sequence rather than a pivot screenshot
Build the D1 map, 08:00 ATR snapshot and first-contact state from data available at those historical moments. Retain the broker calendar and price sides. An indicator that paints today’s final high/low into an earlier daily map introduces future data. M15 OHLC can identify A/B conditions but cannot prove the first-quote deadline, entry spread, short exits or which exit fired first.
Use ordered Bid/Ask ticks for execution and record missing or stale quotes. Include commissions, spread changes, rejected requests, adverse fills and timed exits. Split development, chronological out-of-sample assessment and forward observation. Testing many ATR multipliers, sessions, confirmation delays and pivot feeds then publishing only the best curve creates selection bias. Keep a version list, including failed specifications.
Report net expectancy, drawdown, turnover, trade counts and sensitivity to execution assumptions, not just contact frequency. Clustered days and overlapping data are not independent trials. Use the MT4 backtesting guide for platform limitations and compare the frozen experiment with subsequent forward records. No minimum sample size or past ratio alone guarantees future profitability.
15. Journal filled, skipped, cancelled and ambiguous cases separately
Save broker day and UTC offset, prior D1 H/L/C and timestamps, P/S1/R1, 08:00 ATR and band, first-contact time, approach close, A/B OHLC, availability times and cancellation reason. Then save the first quote’s timestamp/age/spread, rounded stop/target, r/g/c/a, planning ratio, budget, volume, order response and actual entry/exit fills. Preserve a screenshot as supporting context, not as a replacement for this record.
An invented reconciliation example contains 10 first-contact attempts: 4 invalid A, 2 invalid or expired B, 1 quote/cost skip and 3 fills. Of the 3 fills, 1 reaches target, 1 reaches the stop and 1 closes at the time exit. The categories sum to 10, but neither 3/10 fills nor 1/3 target exits measures a validated win rate. The timed exit may be positive or negative after costs; real per-trade returns are required.
Keep incomplete tick histories and ambiguous exit order outside resolved performance totals, with their own counts and sensitivity analysis. Do not count the same quote skip twice under spread and cost reasons; record the first failure and optional additional flags. If a later editorial or coding correction changes a rule, version the experiment and rerun the affected sample instead of silently combining incompatible histories.
16. A practical review before enabling the experiment
Before 08:00, verify the broker calendar and previous D1 data. At 08:00, freeze the map, completed-bar ATR, bands, obstacles, parameters and risk budget. During the window, consume the first contact at each level and accept only A followed immediately by valid B. Before sending, verify the first quote, deadline, tick-rounded prices, net room, protection, size, margin and unused daily allowance.
After entry, reconcile the actual fill and protection, apply the exception rule if necessary, then retain the fixed exits and scheduled close. Afterwards, include skips, rejected orders, losses, timed exits and unresolved cases in the journal. The purpose is a reproducible decision record that can reveal an unprofitable hypothesis as readily as a promising one.
For the underlying references, revisit the support/resistance drawing guide and the round-number guide. Their overlapping lines are not independent probability votes. A later breakout-and-retest strategy will need its own completed-bar sequence; it should not be blended into this bounce rule after the fact.
Frequently asked questions
What is a daily pivot point bounce strategy?
It tests a return from calculated S1 or R1 toward the central pivot after rejection and a separate completed confirmation candle. This version specifies broker-day inputs, M15 timing, first-contact attempts, executable quotes, costs and risk. A touch alone is not an entry and no profitable result is established here.
Can I buy as soon as price touches S1?
Not under these rules. The first contact must form a valid rejection A, then the immediately following B must close above A.high without a new lower low or a prior target touch. Only the first subsequent quote is assessed. A limit order resting at S1 is a different strategy.
Which pivot formula and time zone should I use?
This baseline uses Traditional P = (H + L + C) / 3, S1 = 2P − H and R1 = 2P − L from the latest completed broker D1 Bid bar. Retain that broker’s calendar and date-specific UTC offset. Do not mix a New York-close map with another feed’s daily candles.
Why does the SELL stop include a spread allowance?
The rejection high is observed on Bid, whereas a short stop is checked on Ask. This version adds a fixed 1-pip buffer and the predeclared 2-pip spread allowance to A.high. The allowance does not recreate historical Ask or guarantee protection against later spread widening.
What if confirmation arrives with too little reward left?
Skip when the conservative ratio (g − a − c) / (r + a + c) is below 1.5. The worked higher-price BUY variant has 17-pip risk and 28-pip reward before extra allowance and commission; its planning ratio is about 1.41. Do not move the target beyond P or use an earlier unavailable fill to rescue it.
Is this pivot-bounce strategy profitable?
No profitability is demonstrated. The examples and 10-attempt ledger are invented. A useful assessment needs fixed rules, ordered Bid/Ask data, realistic costs, complete rejection/skip records, chronological out-of-sample work and forward observation. Stops and a 0.5% planning budget do not guarantee a maximum loss.
Sources and execution notes
Platform documentation is used for pivot formulas and quote-side execution conventions. The signal rules, parameters, price paths, sizing examples and ledger are original fictional teaching exercises; the sources do not validate their profitability.
Educational content. All diagram prices, trade examples and ledger counts are fictional. Forex trading involves risk; protective stops and planning budgets do not guarantee a maximum loss.