Cross-Market Leads and the Indicative Openeric

This is part five of the ASX order book series. The previous post argued that the cause of a gap matters more than its shape. One of the most reliable ways to judge whether a gap is justified is to compare it against what the rest of the world did overnight.

Every journal I read had a pre-market routine, and every routine started in the same place: what happened while Australia was asleep.

The ASX opens last

By the time the ASX opens at 10am Sydney time, a lot of relevant information has already been priced somewhere else:

• US equity markets have closed.
• ASX 200 futures (the SPI) have traded overnight and in the early morning.
• Commodity futures and spot markets (iron ore, gold, oil, copper, coal and others) have moved.
• Dual-listed and cross-listed stocks have traded on their other exchanges.
• Sector ETFs and peers overseas (uranium miners, lithium producers, coal companies) have repriced.
• The Australian dollar has moved, which matters for exporters and for commodities priced in US dollars.

This means the ASX open is, in a sense, a catch-up event. The question for an intraday trader is whether the catch-up in the indicative price is too much, too little, or about right.

The leads the traders actually used

From lead to expected gap

The simplest useful model is to ask: given the overnight move in the relevant lead, how far should this stock open?

Copy expected gap ≈ beta × lead move
mispricing   = indicative gap - expected gap

Beta here is the stock's typical sensitivity to that lead, estimated from history. A large iron ore producer might move roughly in line with the iron ore price; a smaller, higher-cost producer might move more. A gold developer might be far more sensitive to the gold price than a large producer.

Here's a sketch of how you might compute it, given daily history.

Copy import pandas as pd

def lead_beta(stock_ret: pd.Series, lead_ret: pd.Series, window: int = 120) -> float:
    """Rolling sensitivity of a stock's open-to-prior-close move to an overnight lead."""
    df = pd.concat([stock_ret, lead_ret], axis=1).dropna().tail(window)
    cov = df.cov().iloc[0, 1]
    return cov / df.iloc[:, 1].var()

def mispricing(indicative_gap: float, lead_move: float, beta: float) -> float:
    expected = beta * lead_move
    return indicative_gap - expected

This isn't a trading system. It's a lens. A large positive mispricing means the stock is indicated higher than its lead justifies (a fade candidate). A large negative one means it's indicated lower (a bounce candidate). Then you go back to the book: is there protection for the trade?

Comparing against peers

When there's no clean lead, or when the lead itself is noisy, the traders compared each stock against its sector peers. This is relative value at the open.

Here's a synthetic example from a hypothetical morning where gold rallied overnight.

Gold C is indicated far above its peers with no stock-specific news. Unless there's something you don't know, the auction is overpaying. Gold D is indicated down on a morning when its sector is up. That's either news you need to find, or a mechanical gap (a lone seller, for example) that's worth a closer look.

The same idea works across a basket during the session, too. When a sector rallies, the traders often looked for the laggard: the name with the least move, on the theory that it would catch up as buyers rotated through the group. Several journals described sizing up harder in the name that lagged.

Dual listings: the cleanest reference

Some ASX companies also trade in London, New York or Toronto, either directly or through depositary receipts. When the other listing has traded overnight, it gives you about the most direct estimate possible of where the ASX line should open, after adjusting for currency and any ratio.

If the foreign listing closed down meaningfully and the ASX line is indicated up, something is off. Picture a large cap indicated higher in Sydney while its overseas line and its closest peer both fell overnight, with no company news. Unless you are missing something, the auction is overpaying, and a fade from that open has the odds on its side.

Be careful with dual-listing arbitrage beyond the open, though. Converting between lines takes time and has costs, short borrow may not be available, and "fair value" can stay wrong for longer than an intraday trader can stay in the trade.

Watching the lead during pre-open

A subtle point that came up repeatedly: the lead isn't frozen at the overnight close. Futures and some commodities keep trading into the ASX open. A thesis built on last night's close can be invalidated by a move at 9:55am.

Several traders described pulling a planned match order because the relevant commodity or the futures turned against them during pre-open. Others described the opposite: a gold rally that continued right into the ASX open, making gap-downs in gold stocks clearly overdone.

Where it goes wrong

• Stock-specific news. The lead says up, but the company has announced a downgrade. Always check announcements first.
• Changing betas. A stock's sensitivity to its commodity changes as its business, hedging and balance sheet change.
• Crowding. On big commodity days, everyone is looking at the same comparisons. The obvious mispricing may already be competed away in the auction.
• Index days. On rebalance or expiry days, mechanical flows can swamp the lead.

A morning template

Next, we look at something that makes all of this harder: orders that aren't what they seem.