Find a published ratio or construct an explicit pair
TradingView lists TVC:GOLDSILVER as Gold/Silver Ratio. Open it from Symbol Search for a published reference series. Its public overview confirms the meaning of the ratio but does not document enough input and timing detail to assume that it equals your broker prices or a particular futures pair. For reproducible research, select and record the two underlying feeds yourself.
- TradingView supports division directly in Symbol Search. Select the first full symbol, type / followed by a space, and select the second. A Compare overlay or rebased percent chart does not calculate this quotient.
- Open each leg separately and inspect its description and provider documentation. A familiar XAU or XAG label alone does not prove spot delivery, quote units or identical data timing.
- Keep gold in the numerator. Silver divided by gold is the reciprocal and reverses the relative-strength interpretation.
| Research question | TradingView action | Record before use |
|---|
| Read the published ratio | Search TVC:GOLDSILVER and open its overview and chart | Provider, timeframe and observation time; do not infer its component symbols |
| Compare spot or broker metal quotes | Search gold / US dollar and silver / US dollar separately; choose one provider, then copy both full identifiers into gold identifier / silver identifier | Instrument description, spot or CFD classification, currency, quote unit and session for each leg |
| Inspect continuous COMEX futures | Enter COMEX:GC1! / COMEX:SI1! in Symbol Search | The dated contract behind each continuous series, each roll date and adjustment setting |
| Study specified delivery months | Open the Contracts list on each futures symbol page and select the required dated contracts before composing their quotient | Both delivery months; do not assume the first continuous contracts have matching maturities |
Match currencies and troy-ounce quote units
Use G for gold and S for silver, both in USD per troy ounce, so R = G / S. All prices and outcomes below are invented arithmetic examples, not current quotes. Gold at 2,400 and silver at 30 give R = 80. Unlike a copper-per-pound divided by gold-per-ounce calculation, this comparison uses the same mass unit on both sides.
- CME quotes the standard GC and SI contracts in dollars and cents per troy ounce. A displayed silver price of 30.00 represents 30 dollars, not 30 cents. Verify imported data units instead of applying another factor of 100 automatically.
- If a source uses grams, kilograms or ordinary ounces, normalize both prices to the same mass convention before division. Contract quantity is separate from the per-ounce quote.
- Reject missing, zero or nonpositive prices for this workflow. A zero denominator is undefined; a very small erroneous silver quote can create an extreme ratio without a genuine market event.
| Input convention | Calculation | Result |
|---|
| Both prices in USD per troy ounce | 2,400 / 30 | 80 |
| Silver supplied in cents per troy ounce | 2,400 / (3,000 / 100) | 80; dividing directly by 3,000 would incorrectly give 0.8 |
| Gold in EUR; silver in USD; assume 1 EUR = 1.20 USD | (2,000 × 1.20) / 30 | 80 after contemporaneous currency conversion |
| Both USD prices converted with the same FX rate | (2,400 / 1.20) / (30 / 1.20) | 80; the common conversion cancels |
Use the same observation clock and session
Choose an explicit sampling rule, such as the close of the same completed 15-minute interval during overlapping electronic sessions. Set both charts to the same display timezone, then record the interval end in UTC as well as the exchange timezone. Matching the display timezone helps identify bars; it cannot change when a feed actually updated.
- For COMEX futures, check the ETH/RTH selector on each underlying chart. TradingView documents RTH as a shorter session than ETH. This selector affects intraday data, not the contents of its daily candle. Use the same intended session and verify both legs have a completed bar for the observation.
- Check delayed-data labels, holidays and session breaks separately. If one quote is delayed or stale, compare both at the latest common completed interval or mark the observation unavailable.
- For daily research, record whether each close is a last-trade price or a settlement. Equal date labels do not prove simultaneous prices. Do not pair a spot close with a futures settlement and call it a synchronized spot ratio.
- A stale-silver example: gold moves from 2,400 to 2,448 while the last silver quote still reads 30. The displayed ratio becomes 81.6, or +2%. If silver at the matched time is actually 30.6, the matched ratio is still 80. The earlier comparison contained a timing mismatch.
- Use the Data Window to copy both leg closes for a chosen bar and independently divide them. TradingView warns that spread bars can differ after reload because live bars use ticks while historical bars use minute data. Synthetic highs and lows are not proof of executable paired prices.
Decompose the ratio return into its two legs
For positive matched prices, define rG = G1 / G0 - 1 and rS = S1 / S0 - 1. The exact ratio return is rR = (1 + rG) / (1 + rS) - 1 = (rG - rS) / (1 + rS). Subtracting the two percentage returns is only an approximation. Each scenario starts at G0 = 2,400, S0 = 30 and R0 = 80.
- In the falling-price case, the return difference is 4.50 percentage points, but the ratio return is 0.045 / 0.90 = 5%. A long gold position still loses money despite the rising ratio.
- An exact ratio-point attribution can change gold first: (2,268 - 2,400) / 30 = -4.4 points. Then change silver: 2,268 × (1 / 27 - 1 / 30) = +8.4 points. The total is +4 points, taking 80 to 84.
- That sequential attribution assigns the interaction according to the chosen order. Reversing the order changes the two contributions but preserves the total. State the convention instead of presenting the components as unique causal estimates.
- A high ratio is a relative price level. It supplies neither a required reversion date nor a reason to buy silver outright. Retain the single-metal charts and a separate test of any proposed entry and exit rule.
| Hypothetical path | Gold endpoint and return | Silver endpoint and return | Ratio endpoint and return |
|---|
| Gold rises; silver flat | 2,520; +5.00% | 30.00; 0.00% | 84; +5.00% |
| Both fall; silver falls faster | 2,268; -5.50% | 27.00; -10.00% | 84; +5.00% |
| Both rise; gold rises faster | 2,646; +10.25% | 31.50; +5.00% | 84; +5.00% |
Convert futures prices into exposure before evaluating a spread
CME standard Gold futures represent 100 troy ounces per contract; standard Silver futures represent 5,000. At the assumed starting prices, one GC has $240,000 notional and one SI has $150,000. Long one GC and short one SI therefore starts with $90,000 more gold notional. The quote ratio of 80 is not a contract-count ratio.
- For nG long gold contracts and nS short silver contracts, gross P&L = nG × 100 × (G1 - G0) - nS × 5,000 × (S1 - S0). The table holds contract counts fixed and assumes no roll. All three ratio returns are +5%, yet dollar P&L differs.
- Five and eight are arithmetic quantities for this example, not recommended position sizes. Equal starting notionals do not make volatility, margin requirements or losses equal; exposure also drifts as prices move.
- Deduct both legs' commissions, bid-ask spreads, slippage and any roll costs before assessing results. Report return against a declared capital basis; notional exposure and posted margin are different denominators.
- Micro, smaller and broker contracts have their own multipliers. Read the exact traded instrument specification before entering quantities or P&L in a journal.
| Fixed hypothetical position | Starting exposure | What it measures |
|---|
| One GC / one SI notional ratio | (100 × 2,400) / (5,000 × 30) = 1.6 | Contract values, not the per-ounce ratio of 80 |
| Long five GC; short eight SI | +$1,200,000 gold; -$1,200,000 silver | Equal starting dollar notionals, without rebalancing |
| First path: gold rises; silver flat | 500 × 120 - 40,000 × 0 = +$60,000 | Gross two-leg P&L |
| Second path: both fall | 500 × (-132) - 40,000 × (-3) = +$54,000 | Gross two-leg P&L |
| Third path: both rise | 500 × 246 - 40,000 × 1.5 = +$63,000 | Gross two-leg P&L |
Separate contract switching from a metal-price move
GC1! and SI1! each join successive contracts. Their switches need not occur together or select the same delivery month. On each underlying chart, enable Contract Switching in chart settings under Events and open the switch icon to record the old and new contracts. Inspect both dated contracts at the same timestamp before explaining a ratio jump.
- The old and new prices in this table are assumed to coexist at one instant. No individual contract moved. The apparent ratio changes come entirely from selecting different maturities.
- TradingView back-adjustment changes earlier contracts using the new-minus-old close difference near a switch. Record B-ADJ separately for gold and silver, where supported. Applying it to both legs does not preserve past ratios.
- For example, assume the adjustment offsets are +24 for gold and +0.60 for silver. An earlier observed pair of 2,200 and 25 had ratio 88. Adding those offsets changes its displayed ratio to 2,224 / 25.60 = 86.875. That revised history is not the originally observed price pair.
- For a trade study, use the actual dated contracts, explicit roll transactions and their costs. Preserve the original observations if your hypothesis used historical ratio levels. Continuous-chart returns alone cannot reconstruct those executions.
| Fixed-price roll illustration | Selected prices | Ratio and change from 80 |
|---|
| Before either switch | Old gold 2,400 / old silver 30.00 | 80.000000; reference |
| Only gold switches | New gold 2,424 / old silver 30.00 | 80.800000; +1.0000% |
| Only silver switches | Old gold 2,400 / new silver 30.60 | 78.431373; -1.9608% |
| Both switch | New gold 2,424 / new silver 30.60 | 79.215686; -0.9804% |
Keep a reproducible research record and review the price charts
Write the hypothesis before the outcome: the exact ratio definition, observation rule, candidate condition and what would invalidate it. A research record should retain both full symbols, provider classifications, currencies, units, UTC timestamps, session settings, delayed-data status, contract months, B-ADJ settings, prices and the independently calculated ratio.
- Pineify Trading Journal has session notes, manual trade records, tags and closing notes. Keep the setup definition and screenshot filenames or URLs in notes, then record actual or clearly labeled simulated trades and their results. Keep original image files in your own storage; this workflow does not assume automatic TradingView synchronization.
- If no trade occurred, retain the observation in session notes or a research sheet. Do not create a fictional fill to make a ratio observation appear in performance statistics.
- AI Chart Analysis accepts a chart image and its upload guidance asks for a clear candlestick chart. Keep the ratio line chart as research context; for image review, provide a separate readable gold or silver price candlestick chart with symbol, timeframe and axes visible. Review each leg separately rather than assuming that the tool validates a synthetic ratio line.
- Compare any image-based commentary with the recorded numerical inputs. A screenshot cannot certify feed synchronization, contract multipliers, roll adjustments or an executable two-leg return. This workflow does not automatically place trades.
