Real-time data feed lag, RAM, and network latency monitoring on any chart — know if your platform is healthy before you click buy. Color-coded warnings for delayed feeds, throttled measurements, and visual veto power on impulsive trades.
Section 01 — Overview
BWT System Performance displays a set of system health metrics as a pinned overlay on any chart: computer time, chart time, the difference between them (data feed lag), available RAM in megabytes, and network latency in milliseconds. This information is critical for traders who need to know whether their chart data is arriving in real time or whether a feed delay is distorting the picture they are trading from. A two-second lag is an eternity in a fast market — and is invisible without an explicit display.
The lag reading — the difference between computer time and the timestamp on the last market data tick — is color-coded: neutral when under 0.5 seconds, yellow when between 0.5 and 1 second (warning), and red when exceeding 1 second (critical). RAM and network latency are measured on throttled intervals to avoid performance impact: RAM refreshes every 5 seconds and network latency pings every 60 seconds using a public DNS server as the reference endpoint. The throttling is deliberate — the indicator should never become the source of the performance problem it's monitoring.
Treat this indicator as infrastructure, not analysis. It does not tell you when to buy or sell. It tells you whether the data driving every other decision on your chart is trustworthy. A trader with no edge who refuses to enter on lagged data outperforms a trader with great signals who ignores the lag warning. The Preview label reflects that additional metrics and configuration options are planned for future releases.
Screenshot
Image provided by you
Section 02 — Key Concepts
A working glossary of every reading the indicator displays. Knowing what each number represents — and what action to take when it changes — is the difference between System Performance being decoration and being the veto on your worst trades.
The current time according to your local system clock. This is the reference point against which all other time-based readings are compared. If your computer's clock is incorrect (NTP not synced), every lag reading will be wrong — so verify your system clock matches authoritative time before relying on the lag display.
The timestamp on the last market data tick received by NinjaTrader for this instrument. In a live, well-functioning feed, chart time should be within a few hundred milliseconds of computer time. When chart time falls behind, you are looking at stale data — the candle on screen no longer represents what the market is doing right now.
The difference between computer time and chart time — the single most important reading in the indicator. Lag is the gap between reality and the chart; a 1-second lag means a 1-second-old picture of the market. In a fast tape, that 1 second can be 4–5 ticks of price movement. Lag readings are color-coded so the warning state is unmissable.
Free system memory in megabytes, refreshed every 5 seconds. NinjaTrader memory consumption grows over a long session — chart data, indicator state, order history. When available RAM drops too low, NinjaTrader's responsiveness suffers: clicks lag, charts redraw slowly, and order entry becomes unreliable. The fix is to restart NT before performance collapses.
Round-trip time in milliseconds to a public DNS endpoint (Google DNS, 8.8.8.8), refreshed every 60 seconds. This measures the health of your local network connection — not the broker connection specifically, but a reasonable proxy. Sustained high latency points to ISP issues, WiFi degradation, or a saturated local network. Wired connections almost always outperform wireless.
When lag reaches 500 milliseconds (0.5 seconds), the lag display turns yellow. This is the "pause new entries" threshold — half a second of feed delay is enough that a fast move has already happened by the time you click. Existing positions can still be managed, but no new entries until lag returns to neutral.
When lag exceeds 1 second, the lag display turns red. This is the "stop trading" threshold — at this point, the chart and the market are far enough out of sync that no trading decision is reliable. Pull all working orders, finish managing existing positions to stop or target, and don't re-enter until lag has returned to neutral and stayed there.
RAM and network latency are deliberately not measured every tick. Querying RAM on every tick would itself consume measurable CPU; pinging an external server every tick would generate thousands of network requests. The 5-second / 60-second cadence is the lowest-overhead schedule that still surfaces meaningful changes — the indicator never becomes the cause of the performance problem it's trying to detect.
Section 03 — Daily Routine
A repeatable session sequence for using System Performance as an active part of your trading discipline. The indicator only adds value if you've decided in advance what you'll do at each warning level.
Section 04 — Parameters
Minimal configuration by design — the indicator should be deployable on every chart with a single click and a glance to confirm the readout location. The defaults are tuned for the most common use case: pinned bottom-left, neutral text, lag warnings color-coded.
| Parameter | Default | Description |
|---|---|---|
| Text Position | Bottom Left | Where the overlay appears on the chart: top-left, top-right, bottom-left, bottom-right, or center |
| Font | Courier New 14 | Font face and size for the displayed metric readings — monospaced fonts keep numbers aligned as values change |
| Lag Warning Color | Yellow | Text color applied to the lag reading when it is between 0.5 s and 1 s |
| Lag Critical Color | Red | Text color applied to the lag reading when it exceeds 1 s |
Section 05 — Use Cases
These are the most field-tested ways traders use System Performance. Each one assigns the indicator a specific job in your workflow — making it an active veto on bad trades rather than a passive readout you stop noticing.
A 2-second glance at the readout immediately before clicking buy or sell. If lag is neutral and RAM/latency are within normal ranges, proceed. If anything is yellow or red, the trade is automatically cancelled — no exceptions, no "but my setup is perfect." This is the highest-value use of the indicator: it converts platform health into a binary go / no-go condition that runs every time you commit capital.
Pin System Performance on every chart you trade from for the full session. The indicator's job is to catch lag spikes that occur between trades — sometimes a feed degrades minutes before you click, and your peripheral vision catches the yellow before your conscious attention does. This passive monitoring is most valuable on slower-paced strategies where you might glance at the chart once every several minutes.
Track the RAM trend across the session. NinjaTrader's memory footprint grows over hours of running — historical data caches, indicator state, drawing object accumulation. When RAM drops below the threshold you've identified as causing slowness on your specific machine, restart NT before the slowdown affects an active trade. The exact threshold varies by hardware; learn your machine's tipping point and act ahead of it.
When orders are filling poorly or feeds feel "off," the network latency reading is your first diagnostic data point. A sustained latency reading above 100 ms — when your normal is 20–40 ms — points strongly to a local network issue, not a broker problem. Conversely, normal latency with high lag points to a broker-side feed issue. The combination of metrics narrows the troubleshooting fast.
When lag goes yellow or red while you're in a position, the rules change. You're not pausing entries — you have a live trade that needs management. Switch to "stop and target only" mode: rely on your already-placed protective orders, do not attempt to discretionarily exit on the lagged tape, and do not try to trail manually. The pre-set bracket is exactly what carries you through a degraded feed; manual decisions on bad data are the danger.
Section 06 — Best Practices
A consolidated set of practices for using System Performance as an active discipline tool. Every one of these has cost real traders real money to learn — adopt them up front.
If a chart is one you place orders from, System Performance belongs on it. The indicator costs nothing in CPU and adds a permanent veto layer to every trade. Make it part of your standard chart template so a new chart inherits it automatically — this is the only configuration that ensures you can never forget to add it.
When the lag display is in its neutral / default color, the feed is healthy and entries can proceed normally. This is your green light. If you find yourself rationalizing a trade with anything other than neutral lag, you've already lost the discipline battle — the rule is binary, not negotiable. Neutral or no entry. Period.
When lag turns yellow (0.5–1 second), cancel any pending orders and stop initiating new positions. Existing positions can still be managed via their attached brackets. Yellow typically resolves within seconds; the patience to wait it out costs nothing. The trade you would have taken into a yellow lag is the trade you'll wish you hadn't taken.
Red lag (over 1 second) is the most dangerous state. Stop all trading activity. If you have a position open, let the bracket work — do not try to discretionarily exit on stale data. If you have no position, stay flat. Red typically lasts longer than yellow; if it persists more than 60 seconds, investigate platform connectivity actively (check broker status page, restart connection if needed).
Identify the RAM level at which your specific machine starts feeling sluggish — this is hardware-dependent. For most modern systems, available RAM dropping under 1500 MB is a clear restart signal. Don't wait until the platform is actively unresponsive; schedule the restart for a flat moment between trades. Restarting under duress is much worse than restarting on a calm signal.
A normal residential broadband connection should produce latency in the 15–50 ms range to public DNS. Sustained readings above 100 ms warrant investigation: WiFi degradation, ISP issues, broker-side problems. Don't trade through it — diagnose first. Wired ethernet drops latency dramatically vs. WiFi for most setups; if you're not on a wired connection during live trading, that's the first fix.
The default Bottom Left position keeps the readout out of the price action zone. Choose a corner where the readings are in your peripheral vision — close enough to catch a color change without effort, far enough that they don't compete with candles for visual attention. Test by trading on the chart for an hour: if you find yourself looking at the indicator instead of price, move it.
The indicator's value is precisely that it can stop you from taking a trade you would otherwise take. If you let the readings stop you only when you weren't going to trade anyway, the indicator does nothing. The hard cases — the ones where a great-looking setup appears during a yellow lag — are exactly when the discipline matters. Skipping a great setup because of a yellow reading is the price of paying nothing for the much larger losses you're avoiding.
Lag, RAM, and latency vary by hardware, network, broker, and time of day. Spend a week observing your normal session readings before relying on the warnings. Know what neutral lag looks like for you (typically <100 ms in good conditions), what RAM range is healthy, what your typical latency is. The thresholds in the indicator are general — your personal baseline tells you when something is genuinely off.
System Performance handles the technical health side; BWT ChartNotes handles the trade plan side. Together they form a complete pre-trade checklist on a single chart: is the platform healthy (System Performance), and am I taking a trade that aligns with my plan (ChartNotes)? Two indicators, no alt-tab, both running passively until you need them.
Section 07 — Common Pitfalls
The failure modes are almost entirely behavioral. The indicator does its job perfectly — it shows the readings. The traps below are the ways traders manage to lose money despite having the warnings displayed.
The exact trades the indicator is trying to stop you from taking are the trades you most want to take when you're emotional. Down on the day, chasing a setup, lag turns yellow — and you click anyway. The indicator can't stop you. Only your pre-committed rule can. Decide in advance: yellow = no entry, full stop.
Red lag over 1 second means the chart is reliably wrong. Trading through it isn't risk-taking — it's gambling on what the price actually is, blind. The losses from red-lag trades are often catastrophic because the gap between the chart and reality has already moved against you by the time you see the fill.
Treating the warning thresholds as universal misses the point. Your machine, network, and broker combination has a specific normal range. Without that baseline, you don't know whether 80 ms latency is normal-for-you or a developing problem. Measure for a week, then calibrate.
An indicator showing yellow lag does nothing if you haven't decided in advance what yellow means. The action plan must be specific: "yellow = cancel working orders, do not initiate new entries, manage existing positions only." Vague intentions like "be careful" don't survive contact with a great-looking setup.
A single brief yellow flash often is nothing — momentary network blip. But the cost of treating a yellow as nothing and being wrong is enormous; the cost of treating a yellow as a pause and being too cautious is zero. Asymmetric outcomes warrant asymmetric responses. Always pause on yellow, even if it resolves in a second.
Lag spikes are obvious; RAM decline is silent. Available RAM drops 50 MB an hour over a 6-hour session and you don't notice until the platform feels sluggish. Glance at the RAM reading hourly, not just when something feels wrong. By the time NT feels slow, you've already lost reaction time on every fill.
System Performance is a warning indicator, not a fix. It tells you the platform is degraded; it does not repair the lag, free up RAM, or improve your network. The action is yours: pause, restart, switch to wired. Confusing the warning with the solution is how traders end up watching a red lag display while still trying to trade.
After several days of nothing but neutral readings, traders start to feel the indicator is "in the way" and remove it. Then a feed problem hits during a live trade and they have no warning. The indicator's value is precisely that it does nothing 99% of the time and saves you on the 1% day — keep it on regardless of recency bias.
BWT Precision Indicators require a valid BWT license for NinjaTrader 8. This page is provided for informational and educational purposes only and is not trading advice. Trading futures and other leveraged products involves substantial risk of loss and is not appropriate for all investors. Past performance is not indicative of future results.